10 NHL Draft Day Pick Trades That Backfired (2006-2016)

Sometimes at the NHL Draft, teams will trade two or more lower picks to move up and get their guy.  Other times, teams are simply looking to trade down to add more picks. Here are the 12 worst examples of draft pick trades that happened on Draft day (2006-2016). These are only trades that were draft picks for draft picks (two or more lower picks for a higher pick), no player for pick transactions. Some of these trades included picks from future drafts, so not all of them were drafted on the day they were traded. 

1) 2008 Nashville: Trades #15 (Erik Karlsson) to Ottawa for #18 (Chet Pickard) and #70 (Taylor Beck). David Poile probably thought he was getting great value adding a 3rd round pick to move down 3 spots in the first round. If he had to do it over again, he probably would have drafted the future Norris winner instead of Chet Pickard.

2) 2008 Phoenix: Trades #38 (Roman Josi) to Nashville for #46 (Colby Roback) #76 (Matt Brodeur). The day after trading down from the Erik Karlsson pick, the Preds turned around and traded two picks to move up in the 2nd round to take Roman Josi, who has developed into a great defenseman himself. It's a bit remarkable that these two extremes of good and bad draft day trades involved the same team at the same draft.

3) 2006 Islanders: Trades #71 (Brad Marchand) to Boston for #98 (James Delroy) and #126 (Shane Sims). In most years draft pick value really starts to level out somewhere in the 3rd round depending on the depth of that year's draft class. Acquiring a 4th and a 5th for a 3rd is not a bad gamble, unless the other team hits a home run in that slot. 

4) 2007 St. Louis: Trades #9 (Logan Couture) to San Jose for #13 (Lars Eller), #44 (Aaron Palushaj) and #87 (Ian Schultz). Lars Eller would go on to be a decent 3rd line NHLer with 213 career PTS and counting. But Couture has 424 and that gap isn't going to get any smaller going forward.

5) 2011 Toronto: Trades #30 (Rickard Rakell) and #39 (John Gibson) to Anaheim for #22 (Tyler Biggs). The fact that Biggs was a big strikeout and the Ducks hit a pair of home runs does make this trade look worse now than people probably thought on draft day. The difference in Expected Value between the 22nd pick and the 30th is much smaller than the EV of the 39th, so the Ducks should have been expected to come out ahead, just not to this degree.

6) 2008 LA: Trades #12 (Tyler Myers) to Buffalo for #13 (Colten Teubert) and #74 (Ryan House). Perhaps the worst part of this trade is that both teams passed on Erik Karlsson. The Kings would go on to win 2 Stanley Cups without Tyler Myers, but I'm sure if they could do it over again, they'd rather him than Teubert, who was a complete bust.

7) 2009 Islanders: Trade #16 (Nick Leddy), #77 (Matt Hackett), and #181 (Erik Haula) to Minnesota for #12 (Calvin De Haan). The Islanders got a future NHL player in this trade, but certainly a level or two below Leddy. The fact that Minnesota hit a home run on the 7th round pick does make this trade look worse than it would on draft day. But still, I would generally not recommend trading a 3rd and a 7th to move up 4 spots.

8) 2013 San Jose: Trades #20 (Anthony Mantha) and #58 (Tyler Bertuzzi) to Detroit for #18 (Mirco Mueller). The Sharks did get a player who is currently playing in the NHL, but not at the level where trading a 2nd round pick to move up 2 spots makes sense. Keep the 2nd round pick, draft at 20. Holland did great in this trade, getting two players who are currently in his line-up.

9) 2010 Chicago: Trades #30 (Brock Nelson) to the Islanders for #35 (Ludvig Rensfeldt) and #58 (Kent Simpson). The difference between the 30th and 35th picks is generally less than the value of the 58th pick. That was not the case here, with the Islanders hitting a home run at the 30th slot. 

10) 2014 Minnesota: Trades #79 (Brayden Point) to Tampa for #80 (Louis Belpido) and #204 (Jack Sadek). This trade is only bad because Tampa hit a home run on the 3rd round pick, otherwise it was a reasonable exchange of value. I have to assume this trade took place moments before the pick with Tampa specifically targeting Point. Otherwise why trade up one slot in the 3rd round? Did anyone at the Wild draft table say "hey why don't we draft Brayden Point?"

Drafting Stanley Cup Champions (2004-2016)

How many draft picks become Stanley Cup Champions? Of 2500+ non-goalies drafted from 2004 to 2016, 59 have won the Stanley Cup (2%). How does this sample of the draft pick population compare to the non-Cup Winning picks? Not shockingly, the bulk of this sample comes from Chicago, LA, and Pittsburgh. There may be a "chicken or the egg" dilemma with the data set. Did these teams win because they drafted this way, or did they hit on a few super stars (Malkin, Crosby, Toews, Kane, Kopitar, Doughty), bringing everyone else along for the ride? It may be a little bit of both.

The histogram below shows the total number of Cup Winning Picks by round drafted (goalies not included, or players older than 21).

The shape of the graph is a bit different from draft pick value by round for the non-championship winning population. The 2nd round is closer to the 1st round than it should be based on standard draft pick value, the 5th round beats the 3rd round, and there's a flat slope starting from round 3. The Stanley Cup champions do need to hit on some later round draft picks. The Penguins got Letang and Guentzel in the 3^rd, the Hawks got Hjalmarrson in the 4^th and Shaw in the 5^th, the Kings got Alec Martinez in the 4^th and Muzzin in the 5^th. Even with elite talent at the top of the line-up, you need to hit on some later picks to win the title.

The "chicken or the egg" dilemma applies to the success of the 2nd round picks. Are teams more likely to win the Stanley Cup if they hit on round two, or are the 2nd rounders just more likely to be passengers on the train? Players like Krejci, Lucic, Saad, Voynov, Toffoli, made a significant contribution to trophies, others were mostly passengers and role players. There is a steeper decline in success rate from 2nd to 3rd round among Cup Winning Picks than we see in the non-Cup population. If non-1st round picks are just more likely to be passengers, we should see more 3rd rounders winning Cups. This does suggest there is some truth to the theory that it's important for Cup contenders to hit on 2nd round picks.

Where do they come from?

The chart below shows what percentage of the draft population played at least 10 games in each feeder league in their first year of draft eligibility. Some prospects play in more than one league, so the totals don't necessarily sum to 100%.

The proportion of draft pick sources are comparable between Cup and non-Cup draft picks. There is a slightly larger proportion of Canadian major junior players, and a smaller % drafted from tier II. The NCAA number is that small because of the small number of kids who have finished high school by their NHL draft eligibility. Most of the college bound kids are playing tier II. Does this mean that teams are more likely to succeed in the playoffs if they draft more from CHL than the NCAA and its feeder leagues? It's possible. CHL players tend to turn pro sooner, so the dynasty teams that need young guys to plug holes at a discount, can get help faster if they draft from junior.

We do see the Cup Winning Picks with a higher proportion from European Elite leagues, which doesn't necessarily mean that drafting from those leagues will produce more championships. There are very few players who are good enough to play in the higher level elite leagues in their draft year. Teenagers playing in Elite leagues have very high upside, if you can get them.

The chart below shows the rate at which Cup Winning Picks enter the NHL versus the rest of the draft population.

50% of Cup winning draft picks will play at least 10 NHL regular season games at age 20, and over 70% by age 21. These are the ages where we see the greatest difference in growth (or slope) of the samples above. Players who go on to win Stanley Cups are getting fast tracked to the NHL after entry level slide at a greater rate than other picks.  Does this mean that teams are more likely to win the cup if they fast track their draft picks? Probably not. We're talking about Chicago, Pittsburgh, LA, teams who have won Cups under the salary cap crunch. It's more likely these teams are forced to fast-track their best young players to plug holes in the line-up for cheap.

10 Biggest NHL Draft Busts (2004-2012)

Who were the 10 biggest NHL Draft Busts from 2004-2012? These are all first round picks. Players who fell below the expectations of their draft position like Nail Yakupov could have been included, but at least he earned an NHL job. Instead the true top 10 list should be players most fans have never heard of.

1) Zach Hamill, Boston 2007, 7th overall: Played 20 NHL games and 349 AHL games as a top 10 pick. He was a good playmaker but lacked the size and foot speed to stick in the NHL.

2) Scott Glennie, Dallas 2009, 8th overall: Injuries did play a role in his being a draft bust.

3) Dylan McIlrath, Rangers 2010, 10th overall: He has managed to play 43 NHL games but is still languishing in the minors at age 25.

4) Kyle Beach, Chicago 2008, 11th overall: He was expected to be the next Milan Lucic, but it never materialized.

5) Duncan Siemens, Colorado 2011, 11th overall: Was once described by the Hockey News Future Watch as "the biggest bust in his draft class".

6) AJ Thelen, Minnesota 2004, 12th overall: He holds the distinction of highest draft pick not to sign an entry level contract in the salary cap era.

7) Marek Zagrapan, Buffalo 2005, 13th overall: He scored 127 PTS in the AHL on his 3-year Entry Level Contract with 0 NHL GP. Did not get a 2nd NHL contract. Returned to Europe.

8) Sasha Pokulok, Washington 2005, 14th overall: Pokulok was passed over in his first draft after playing in the tier II Saskatchewan Junior League, only to be taken a year later after playing NCAA. He bounced back and forth from AHL to ECHL on his ELC, eventually leaving for Europe.

9) Ryan O'Marra, Winnipeg 2005, 15th overall: O'Mara did get 33 NHL GP but struggled to produce offense at the professional level.

10) Patrick White, Vancouver 2007, 25th overall: Canucks GM Dave Nonis went off the board to draft this kid several spots before his pre-draft rank. He went directly to Europe after coming out of college and did not get offered an NHL entry level contract.

Predicting 2018 NHL Future Watch Rankings

Every year the Hockey News releases a magazine that ranks the best NHL prospects who are not currently playing in the big leagues. I always buy the first copy I see on newsstands every spring and keep Future Watch rankings in my spreadsheet of draft picks from 2004-2016. I lack detailed knowledge of the process that goes into making their rankings, though it is obvious they conduct dozens of interviews with scouts and are clients of the International Scouting Service.

I'm curious to see if I can predict what the top 75 order will be in 2018 with any reasonable accuracy. There is really no point trying to predict who will be in the individual Team top 10, since that will be mostly made up of the same players printed in their beginning of the season yearbook, with some minor alterations. They recently started ranking their top 100 prospects, but historically it was just 75 so I'll be sticking with that number. The HN excludes players currently playing in the NHL at the time they compile their list, who could have been sent back down by the time I made my list. So there will be some players excluded because they were promoted or demoted, before or after submission deadlines.

I started with a statistical estimate of where prospects would be based on draft position and last year's ranking (if they are not already in the NHL), then moved them up and down based on how they are performing so far this season. Hockey News Prospect "guru" Ryan Kennedy shares significant responsibility for Future Watch rankings, and I do listen to his Prospects Podcast. That does provide me with some indication of where they are leaning on certain players. Also, I don't bother ranking goalie prospects.

For more reading on the subject of Future Watch, check out these posts "Probability of Draft Picks Being Ranked In Future Watch" and "Expected Future Value of Future Watch Prospects".

Expected 2018 Future Watch Top 75 Rankings Order:

Why NHL Teams Should Maximize Draft Picks Entry Level Slide

When NHL teams draft 18-year-old players, they have the option to allow them to continue playing amateur for 2 years before they turn pro. Over those 2 seasons, they are allowed to delay the start of Entry Level contracts. Once a prospect plays over 10 games in a season, his ELC kicks in. Over 90% of all draft picks will maximize their ELS, with just a chosen elite few getting rushed into the pros. The earlier an ELC expires, the sooner the next contract begins. Those prospects will get paid more at a younger age

To answer the question of how much it costs to burn a year of ELS, we'll look at 2 groups. One who started their NHL contracts at 19 (with 1-year of ELS remaining) and played at least 10 games; and those who started at age 20 after maximizing ELS and played at least 10 games. We're not including players who went directly to the NHL at 18, since that includes the most elite tier who will make more money on 2nd contracts because they are better hockey players. We're specifically looking at those who were "NHL ready" at 19 versus those who were "NHL ready" at 20, because we want to minimize the amount that talent disparity skews the results.

On top of salary increases happening one year later, those who burned their ELS at 19 appear to get an extra $500K in salary. For example, group one averaged $2.5M at age 22 and group two averaged $2M at age 23. So not only did group one get to their 2nd contract faster, it was $500K larger once they did. You could argue that's because the group who went to the NHL at 19 were more talented players and deserved more money on their 2nd contract.

The area between the two lines on the graph above represents the extra money paid out. The largest wage gap is at age 22, when group 2 will be in their last year of entry level. This is where teams are losing the greatest value by burning the last year of ELS. You can get a marginally productive player at age 19, or get a very productive player at an Entry Level price tag at age 22. To me this feels like a simple choice. It's possible that turning pro a year early will help a prospect reach a higher level later in his career.

Burning ELS at age 18

The group of players who go immediately to the NHL after being drafted (in their first year of draft eligibility) are a different class of player. Many of these players are good right away and deserve a shot to play in the NHL like Sidney Crosby, Patrick Kane, Auston Matthews, Nathan MacKinnon, John Tavares, Jack Eichel, Patrick Laine, Connor McDavid, etc. The Oilers botched the first year of Leon Draisaitl's ELC at age 18, then got stuck giving him a monster contract at 21.

However, there is also a group of players who were rushed to the NHL at 18 and probably could have used another year or two of amateur, like Jesse Puljujarvi, Noah Hanifin, Jakob Chychrun, Luca Sbisa, Rasmus Ristolainen, Mikhail Grigorenko, etc. Rushing young defensemen especially makes no sense to me. Entry level contracts are generally cheaper than standard contracts, at least when dealing with good players. Would you rather get that discount at age 18/19 or age 21/22 when the player is older, stronger, more experienced?

Ottawa Senators Draft Success

The Ottawa Senators are considered to be one of the better drafting teams in the NHL. There is always a pool of NHL-ready talent in their farm system to replace departing players. It begs the question of whether this success is because they draft smarter than other teams, or are they better at prospect development? The Sens do retain their draft picks longer than most teams, with 51% still being owned by the team at age 23. That's 2nd best in the NHL after the Nashville Predators (the league average at this age is just under 40%).

The chart below shows what % of Ottawa Senators draft picks are still owned by the team at each age (for draft picks from 2004-2016), with a separate line showing the average for the rest of the league. The Sens are more than 10% above the league average for ages 23 and 24; staying above the league avg at almost every age.

The Senators do not attract as many big name free agents as other more desirable destinations. They rely on homegrown talent to succeed. This creates a "chicken or the egg" debate; do the Senators sign more picks because they're better than average, or where there simply more job opportunities due to a lack of incoming free agent talent? We know that they retain ownership of their prospects at a greater rate than the rest of the league, but are they playing more NHL games?

Most NHL Draft picks will be on Entry Level Slide for age 18/19, get a 3-year Entry Level Contract for ages 20/21/22, then get their 2nd contract at age 23. It can be different for college players and Europeans, who can wait beyond age 20 to start their ELC. The Senators success rate with draft picks is very similar to the NHL rate from age 18-22. Where they beat the league average is after ELC. Although it should be pointed out that the Senators are among the NHL leaders in % of draft picks still on ELC at age 23 and 24. They are 5th in the league for % of draft picks on a 2nd contract (or more) by age 25.

The number one source of future NHL talent is the Canadian Hockey League, but only 31% of Sens draft picks played more than 9 GP of Canadian Major Junior in their first year of draft eligibility, which was the lowest % among NHL teams. More than half of their picks played North American Tier II or European junior (ranking near top of the NHL in this category), which are often higher risk lottery tickets. This helps to explain why their pick ownership % is above the league average at age 20,21, and 22 despite them not playing significantly more AHL or NHL games.

From 2004-2015 they only drafted 4 defensemen we could say are good enough to be top 4 D. Erik Karlsson at pick 15, Cody Ceci at pick 15, Thomas Chabot at pick 18, and Andrej Meszaros pick 23. They've had success drafting defensemen in the 15-23 pick range. but whiffed on both D they drafted top 10 in Brian Lee and Jared Cowen. The franchise has mixed in some bad picks with the great picks. Yes, they can make mistakes. It's possible that injuries played some part in the Cowen downward spiral, but history will look back on that as a bad draft pick.

Despite the Senators drafting CHL players at the NHL's lowest rate, many of their biggest draft day steals came from that league; Mark Stone rd 6, Zack Smith rd 3, Mike Hoffman rd 5, and JG Pageau rd 4. Getting two elite talents like Stone and Hoffman near the end of back to back drafts is a remarkable feat comparable to Zetterberg and Datsyuk (who are better players, but drafted 5th round or later in back to back drafts). In the case of Stone, he would have fallen in the draft because of his skating, pushing him to work hard to improve his skating. That may be a win for the development side. Hoffman fits more into the "late bloomer" category and took 5 years to even get mentioned in the Hockey News Future Watch.

The biggest flaw in the Sens drafting is goaltending. For all their success with drafting skaters, the net is a different issue. The Senators produced only one NHL goalie since 2003, and that was Robin Lehner (who is average at best and no longer with the team). None of their top 10 prospects in the last Future Watch was a goalie and Craig Anderson is suddenly getting old fast. They'll need to remedy their lack of quality young goaltending in the system.

In conclusion, yes the Ottawa Senators have been better at drafting than many NHL teams over the last decade. They have hit some big home runs in the later rounds, and are among the best at drafting in the middle of the 1st round. Their ability to develop their own prospects has clearly played some factor in this success, proving to be very patient with bringing their top prospects to the big leagues. I do think a big reason for this draft/development success was Bryan Murray, their former GM who is now deceased. The man helped build a dynasty in Detroit, built the foundation of a Stanley Cup champion in Anaheim, then brought his talents to Ottawa.

Murray deserves a tremendous amount of credit for the system that has been built in Ottawa. It's unclear how successful the team will be at drafting and developing going forward without him. Surely the scouting staff is high quality and it would appear that Dorion shares Murray's slow development philosophy. It's looking like they hit a pair of home runs in the 2017 draft with Formenton (47th) and Batherson (121st) who were both fantastic for a gold medal winning Canadian World Junior team. Both should be in the top 50 of the next NHL Future Watch. There is hope for this organization beyond the possible departure of Erik Karlsson.

Expected Future Value of The Hockey News NHL Future Watch Prospects

Every year the Hockey News magazine puts out a publication called "NHL Future Watch" which ranks the best prospects not currently playing in the NHL. They are all either under contract or a team owns their rights. They rank the top 75 prospects in the whole league, then do a top 10 list for each individual team. Last year they started doing a top 100 list, but since most of the historical data is top 75, we'll stick to that for the purpose of this investigation. If you would like to know the probability of any given draft pick getting ranked in the FW, click here.

What can we expect the Future Watch prospects to earn in future NHL salary?

Expected Salary For Top 75 Ranking

The graph below shows the expected NHL salary at age 23 for each FW rank position (using magazines from 2006-2012). That is the age when most entry level contracts have expired. Each dot does not represent an individual player, but the average for each rank position. It will also only include drafted players, mostly because I don't yet have undrafted players in my prospects spreadsheet. Goalies are also not included.

The logarithmic trendline fits to a 0.65 R^2, which is a respectable correlation between Future Watch rank and future salary. That line crosses the million-dollar threshold near the 65th spot in the rankings. If you look at a chart with the expected future value of the first 50 picks in the draft, it would look very similar to the chart above, but with a steeper decline. If I were to do a rival Future Watch publication and just listed the highest draft picks not currently playing in the NHL, I could probably replicate these results. I believe it would be less accurate, but the scatter plot would have a similar shape.

It's worth noting that the #1 overall rank in Future Watch does have a lower expected future salary than some of the ranks following it, which is thanks to Nikita Filatov playing back in Russia at age 23 and registering a zero in the NHL salary column. None of the other top 10 rank positions had a zero.

Expected Salary By Team Rank

These are only for players who ranked in a team's top 10 prospects but did not make the top 75 list. In most cases, players ranked in the top 2 spots on individual team rankings will also be on the top 75, so we start at the Expected Value of the #3 spot.

The linear trendline fit with a 0.78 R^2, which is another strong correlation. If there were no "future value" pattern to the Hockey News prospect rankings, we would see a flat line with a slope and R^2 close to zero. Choosing a single linear trendline is a bit deceptive; it would fit the data better to have two different lines connecting at rank=6, with a steeper slope for the first half and a flatter slope for the back half (unfortunately MS Excel doesn't offer that option). There is no difference between the 7th position and the 10th position. If a prospect is ranked in the top 5 for an individual team, he can expect a higher future NHL salary than someone ranked in the bottom 5; where there is very little difference between the ranks.

The Hockey News does a great job with this publication. Even if I could mathematically predict what the ranks are likely to be every year, they write detailed insightful write-ups for every prospect after doing interviews with dozens of scouts. That's the part that's hard to replicate.

The Road From ECHL to NHL

If an NHL draft pick ends up in the ECHL after they turn pro, what is the probability that they will eventually play in the NHL? It is difficult for prospects to climb out of that hole and make it to the big show, but it's not impossible. Most draft picks turn pro at age 20 and complete their entry level contracts at age 22 if they are lucky enough to even get signed. Most NHL-bound players who play ECHL will do so on their Entry Level Contracts. There are also some undrafted free agents who sign in the ECHL and eventually climb the professional ladder into the NHL, like Alex Burrows. Unfortunately, I don't have a spreadsheet complete with undrafted free agent ECHL stats to determine NHL probability. We'll have to stick to drafted players for this post.

The chart below shows 2 different groups, players drafted from 2004-2012 who played >9 GP in the ECHL at age 20, 21, 22, 23, and those who play >9 AHL GP at each age. Players who played in the ECHL and AHL during the 2013 work stoppage are not included in the sample.

The first thing that's immediately clear is if your goal is to play at least one NHL game by age 25, your chances are far better if you play in the AHL instead of the ECHL, and by a wide margin. However, the ECHL is not a graveyard league for NHL careers. If a draft pick plays at least 1 ECHL games at age 20, his probability of eventually playing an NHL game is 30%; but those playing ECHL games at age 23, the probability of playing NHL games drops all the way down to 5%. On the other hand, AHL players have a decline of 68% to 57% for age 20 to 23.

The older a draft pick is when he plays in the ECHL, the smaller his probability of ever playing in the NHL, which gets close to zero if the player plays low-pro after his Entry Level Contract. The best time to play ECHL is first year pro, but even then, your probability of future success is at least doubled if you stay in the AHL; a far superior league for prospect development. If any draft picks get the phone call to report to the ECHL, it's not the end of the world, but the longer you're there, the darker the outlook. Make the most of the weaker competition and try to put up great numbers in a hopefully small sample.

Probability of NHL Draft Picks Being Ranked in The Hockey News Future Watch

What's the probability that any given player selected in the NHL entry draft will be ranked in the annual Future Watch magazine released by the Hockey News? They rank the top 75 prospects in the whole league, then do a top 10 list for each team. Everybody who appears on the top 75 list will also appear on the top ten list for individual teams. The magazine excludes players who are currently playing in the NHL; so the rank represents the best prospects not playing in the big show that are owned by NHL teams.

The graph below shows the probabilities by pick for age 18, which will be the first season after being drafted. The blue line is being ranked in a team's top 10, and the orange line represents the top 75. Picks are bundled into groups of 10, with the dots representing the average for all the 10 picks preceding it, the continuous lines are just for visual effect.

A 1st round pick is almost 100% guaranteed to be ranked in his team's top 10 the season after being drafted. Players drafted from pick 20 to 30 will have a 60% chance of appearing on the top 75 list, while those drafted 30-40 will have a 20% chance. The evaluation system used by the Hockey News does show a clear preference for ranking 1st round picks in the top 75. If you looked at Expected Value metrics for NHL draft picks, there is not a huge difference between picks 20-30 and 30-40; but certainly looking at how the Hockey News values being a 1st round pick, there is a huge difference.

The graph for age 20 is similar to age 18, but with one key difference. If you are a top 20 draft pick and are not already in the NHL by age 20, your probability of getting ranked in the top 75 is significantly smaller than age 18. There's a proportion of the 1st round pick population that's already a bust by age 20, which is represented in the numbers. There are many high draft picks that are playing in the NCAA at age 20, and those players generally don't see their rankings drop unless they are exposed at the college level (like a Patrick White).

I love the Future Watch magazine and my heart always skips a beat when I see one on a news stand for the first time each year. I think they do a fantastic job of ranking prospects. My biggest complaint is the disparity between picks 20-30 and 30-40, that is larger in Future Watch than we observe in real life.

NHL Draft: "The Russian Factor"

There is a belief in the NHL Draft that players chosen from Russia are less likely to come play professional hockey in North America and more likely to return home once they have crossed the Atlantic. This causes some Russian prospects to fall in the draft, which also allows some teams to steal elite talents with later picks. We've seen several examples of elite Russian talent falling in the draft and being stolen in later picks in the salary cap era, like Kuznetzov, Tarasenko, Kucherov, Cherepanov, etc. They all should have been top 5 picks.

There is no question that the KHL is the highest level of European pro, certainly when it comes to competing with NHL teams for talent. Russia is one of the top producers of hockey players on the planet, and their pro league is heavily subsidized by Russian oligarchs. This allows KHL teams to offer contracts to players that would not be possible using just hockey related revenues. The Swedish Elite league can't afford to offer its players NHL sized contracts. Because Russia is able to compete with the NHL for players, the prospects are less likely to play in America.

How do we measure "the Russian Factor"? It's difficult to measure exactly how much individual prospects are falling the draft due to their country of origin. You could look at Central Scouting rankings for all prospects prior to the draft and see if Russians are being selected below where the scouts rank them (that'll have to be a future post).  This post is not attempting to answer exaclty how much Russian players are falling in the draft due to country of origin. It's also possible that the "Russian Factor" is being factored into scouting rankings, that scouts may downgrade their ranking of prospects they deem less likely to cross the ocean.

The easiest way to measure "the Russian Factor" is to look at the statistics that's causing it. What % of draft picks (from 2004-2013) from Russia are playing >9 GP of North American Pro (NHL, AHL, or ECHL) at each age; versus the rest of Europe and North Americans.

There is clear evidence that Russian players do have a greater tendency to play in Europe than do players from the other major European hockey producers. There is a valid reason why Russian players slip in the draft. By age 24. only 20% of Russian draft picks are playing >9 GP of North American professional hockey, versus 62% for North American players. Most Draft picks from the CHL have a 2 year deadline to sign a contract, while NCAA picks can stay in school and delay their contract deadline. Russians hit their peak of North American professional participation at age 20, then only gets smaller as players get older.

I believe the part of the draft where the "Russian Factor" has its biggest influence is in the first round. General Managers can take incredible amounts of grief for bad picks in the first round, which is by far the most important round, especially the top half. Hockey might have the most "top heavy" entry draft talent of all the professional sports. That's why players with perceived risks and elite talent can fall in the draft order, as teams have a tendency to go with safer selections.

Looking at the North American participation rate for 1st round picks, we can see where this "Russian Factor" is most valid. Just over 60% of Russian first round picks will be playing NA pro at age 23, when most entry level contracts have expired. Over the age of 21, 1st round Russian picks are more than 20% less likely to play NA pro. It does depend on the risk tolerance of the individual GM.

Russia and Sweden produce the highest proportion of elite level talent, with 7% of draft picks earning over $4M NHL salary by age 23 (for Russians those are mostly forwards and for Sweden mostly defensemen). If we think of draft picks like lottery tickets, the Russian tickets have smaller probability of winning a small prize, but more large jackpots.

What about the Russian players who come to North America prior to the draft? Below is a graph comparing CHL Russians with those drafted from Europe. The grey line is North American CHL players drafted to the NHL.

Russians who are playing in the CHL at the time they are drafted are significantly more likely to play North American pro than their countrymen who are drafted from Europe. That is encouraging and gives teams less risk in drafting Russians who have already crossed the ocean, however they will still play NAP at a lesser rate than North American NHL draft picks from the CHL. Crossing the pond when eligible for the draft can help Russians increase their draft stock, but it does not eliminate the "Russian Factor" risk altogether.

Probability of Getting a Top 4 Defenseman or Top 6 Forward in the NHL Draft

NHL Teams looking to find bottom line forwards or bottom pairing defensemen have a wide assortment of journeymen players and waiver wire fodder available on the open market to fill any holes. The holes that get harder to fill via free agency and waiver wire are top 6 forwards and top 4 defensemen (which in this case will be defined as a forward averaging greater than 16m of ice time in the NHL per game, and greater than 20m for defense). These are the assets we covet the most in the draft. So what is the probability of acquiring either of these two asset-types at any given draft pick?

On the graphs below, draft picks are bundled into groups of ten picks, with the probability representing all seasons over age 21 for all picks. Picks are bundled into groups of 10, so each dot you see on a vertical gridline represents the 10 picks before that number. The continuous line is just for visual effect.

 If you draft a forward in the first 10 picks of the draft, you're more likely to get a top line player than if you're drafting defensemen. Forwards may have a higher success rate in the early picks, but it's still the most likely you'll be to hit the jackpot on a defenseman. There is only a 50% chance that your defense drafted in the top 10 picks will develop into a top 4 D, and yet it still offers the highest probability of getting one of these highly coveted assets. It creates a bit of a paradox. If you're going to need a top pairing D anytime soon, by the time you get beyond the 20th pick in the draft, the probability falls to the 10% range.

The graph above came from a sample of draft picks from 2004-2013. The graphs below are a count of the players who played in 2016/17 who played at least 20 GP and averaged over 16m for forwards and 20m for defense and where they were drafted. Also note that there were several undrafted players playing top line roles, but that's a subject of a different blog post.

The probability of getting a 16m forward in picks 1-10 is significantly higher (63%) than picks 11-30 (28%ish). The drop off is steep once you get out of the elite tier of forward prospects. The probability drops down to 10% by the 2nd round, and under 5% by the 3rd round. There are more forwards drafted in the top 10 picks than any other "ten pick bundle" in the draft, at 69% of all non-goalies drafted. Then from picks 11-20, defensemen are drafted more often and have a higher success rate. The middle of the first round has the highest proportion of defensemen drafted (40% of non-goalies) than any other range in the top half of the draft.

 The higher probability of striking out on defensemen in the first round leads to a preference for drafting forwards with top picks. It's slightly easier to hit home runs on defensemen in the later rounds, but that doesn't mean it will produce results. We've seen some cases of future Norris Trophy contenders drafted after the 1st round like Shea Weber, Kris Letang, PK Subban, Roman Josi, etc. It's very difficult to hit a lottery ticket on top 4 D outside of the 1st round of the draft. It's not impossible, just improbable.

The Detroit Red Wings drafted a defenseman in the 1st round in 2007 (Brendan Smith) and 2016 (Dennis Cholowski), and none in between. Then in between those two picks were the retirements of Nick Lidstrom and Brian Rafalski. Over those 8 drafts they drafted 16 defensemen and ended up with 2 third pairing guys. That's what can happen if you rely on the later rounds to fill your top pairings.  The Red Wings haven't been able to draft a top 4 defenseman in over a decade and now find themselves mired in mediocrity.

If you draft forwards in the top half of the first round you are less likely to strike-out, but it’s also your best chance of getting a top pairing defenseman, which are incredibly difficult to acquire. There is a risk-reward quotient to consider based on the needs of your depth chart., but you'll take heat if the pick is a flop. Do you play it safe or roll the dice? That's why they get the big bucks...

Expected Future Value of NHL Draft Picks From Tier II Leagues

What Tier II feeder leagues for the NHL Draft produce the most future value? These are the North American amateur leagues that do not ruin NCAA eligibility, and produced at least 15 NHL non-goalie draft picks from 2004-2016. There are 7 such sources; the US Development Program, US High School, USHL, British Columbia Hockey League, Ontario Junior Hockey League, Alberta Junior Hockey League, and  other Canadian Junior Hockey Leagues.These are the players who were playing tier II in their first NHL draft year, not necessarily the team they were playing on at the time they were drafted.

The Canadian Hockey League (CHL) “Tier 1” Major Junior produces over 40% of all NHL Draft picks, but playing in that league destroys NCAA eligibility (which in my opinion is an absolutely ridiculous rule bordering on criminal that should be challenged in court). So for kids who would like to pursue a college scholarship, they are forced into other less competitive leagues.

United States High School Hockey produced by far the greatest number of draft picks from Tier II with 186 (59% of those coming from Minnesota and Massachusetts), but also produced the greatest number of “strikeouts”. Roughly 50% of all players drafted to the NHL from USHS will never even get an Entry Level contract, compared to 20% from the USDP and 24% from the CHL.

The level of competition in the CHL is substantially higher than all these tier II leagues, so it’s easier to forecast how players from there will develop into professionals. Tier II is always a gamble. There are so many kids who will dominate high school hockey, but disappear against better competition. Many such draft picks will get exposed at the NCAA level and never get an NHL contract, which is why USHS players will get drafted more frequently in the later rounds when teams are willing to take on more risk for the potential of a big reward.

The BCHL is easily the most successful league of all the Canadian Tier II Junior leagues at producing future NHL players. With recent picks like Tyson Jost and Dante Fabbro having lucrative futures ahead of them, the BCHL value will continue to grow. The league steals players from the CHL at a greater rate than other provinces; the reason for which is open to speculation. The league does attract a significant proportion of prospects from outside the province. There is also a greater proportion of kids in BC that would rather follow the NCAA path than Major Junior.

As someone who lives in BC, I’d like to think it’s because we’re smarter out here, but that’s not it. There’s a significant number of Major Junior players who get their educations at Canadian Universities paid for when their junior careers are finished. You can play CHL and still get a quality post-secondary education, but for some reason in British Columbia they place a greater priority on American educations. For that I’ve got no explanation. I was born in Ontario anyway…

Cale Makar was just drafted from the Alberta league last year, but that league rarely poaches CHL talent. Most Alberta kids who are eligible for the CHL will play in the CHL. Unless Makar inspires more youngsters to choose AJHL over CHL, I would not expect the historical trend to change. Scouts won’t suddenly show up to AJHL games in greater numbers to find the next Makar. He’s the exception.

It will always be a higher risk to draft players that you have not seen play against the best competition possible. Most of the USDP players will be scouted at the U18 World tournament, so scouts can watch them play very competitive games in a small sample. Unless other tier II players get an opportunity to compete on the international stage, there will always be more risk drafting them than CHLers. You’ve got to be sure if you’re going to burn a top 20 pick on a risky asset.

Expected NHL Salary of World Juniors Draft Picks

The World Junior tournament is upon us again. NHL fans get an opportunity to watch their team's draft picks perform on the world stage. What future salary can we expect the players from each country to earn in the NHL? This is estimated using the cap hit from NHL Numbers for players who played in the WJC at any age and were drafted to the NHL at any point under the age of 21. Ergo all the players were taken from the pool of NHL draft picks from 2004 to 2011 who played at least 1 game in the tournament.

Team Canada players have by far the highest expected NHL salary of any country at age 23 with $1.8M. The high volume of players produced by Canada makes for very competitive tryouts with the national team. Roughly 50% of Team Canada draft picks will earn over $1M at age 23 and 14% will earn over $4M. Generally, the majority of Team Canada's roster has already been drafted prior to their first appearance in the Tournament. NHL Teams don't often get to draft players who played on Team Canada and instead hope their draft picks make the tournament after having already been drafted. Players who have a productive tournament can dramatically increase their market value.

You may notice the E[V] for Finland is very low. That country has come on strong in recent tournaments, but that bumper crop which started to appear in the 2012 draft has not yet turned 23 years old so are not included in the sample. From 2004 to 2011, there were 5 non-goalie Fins drafted in the first round; from 2012-2016, there were 11. The recent wave of elite Fins does not factor into any of the graphs below or above.

Furthermore, 33% of these draft picks who played for Team Finland would not go on to play any games of North American pro (NHL AHL ECHL) by age 23. It's unclear if the reason is that they didn't want to leave Europe, or weren't good enough to get a contract. Probably a little bit of both. During this time period, Finland was much better known for producing elite goaltending talent, which started to change with the arrival of Teravainen, Ganlund, Barkov, and Ristolainen.

The graph above is expected salary at the age of 23. The next graph takes the total sum of all NHL earnings for players from 2004-2017 before the age of 28. The graph above is an average at a specific age, the graph below is the sum of all ages (up to and including 27). As you can see, Sweden and USA will produce similar salaried players on average, but the Americans produce a greater volume of draft picks.

Coming To America

At what rate do draft picks from the WJC go on to play North American pro? For Team USA and Team Canada the answer is 100%. 

Russia has a similar NA pro conversion rate as Finland but a much higher Expected future NHL salary. This is partially because the Russians produced more high-end talent than Finland over the sample (such as Ovechkin, Malkin, Kuznetsov, Tarasenko, etc). We have seen some of the elite Russian talent fall in the draft because of the "Russian factor" whereby NHL teams are afraid to draft Russians too high on fears they won't leave the KHL. In reality, they are converting players to North American pro at a similar rate to other European countries.

Swedish players will stay in Europe at a similar rate to Russians for ages 18, 19, and 20. By age 21 that starts to change, with 71% of draft picks from Team Sweden on an NHL entry level contract, versus 46% for Russians. Czechs are by far the most likely to come over to North America to play junior hockey. 73% of Team Czech draft picks will play more than 9 games of CHL at age 18.

These graphs do a good job of illustrating the expected future value of NHL draft picks from different World Junior programs. Canada is easily the best. There has been more discussion recently about the Americans closing the gap, but that is not being shown in the data sample that was used here. If the gap has indeed started to close just in the past few seasons, we won't see that in this data. Those prospects need a chance to play more NHL games and accumulate pro salary data before this type of analysis is possible.

NHL Prospects Who Increased Value Most Following World Junior Tournament

Here is a list of the top 20 NHL draft picks who had the biggest increase in expected future value following a World Junior U20 Championship at 19-years-old from 2006 to 2017. Expected future value is estimated using player rank from the Hockey News Future Watch magazine. The magazine is generally released about 2 months after the tournament. They rank the top 75 prospects in the whole league, and the 10 best on each team, so 300 kids get featured.

These are only players who were NHL draft picks and were eligible for Future Watch as 18-year-olds. That allows us to measure their change in Future Watch rank shortly after the tournament. Players born after September 15 are not included because they are only eligible for one more tournament after being drafted, and I wanted everyone measured by the same criteria.

Oh, and goalies are also not included.

1) Kirill Kaprizov, Russia 2017, Minnesota: This 5th round pick was ranked as Minnesota's 6th best prospect in the 2016 FW, then after 12 PTS in the WJC and a great season in the KHL, Kaprizov jumped up to the 13th best NHL prospect in 2017.

2) Kris Letang Canada 2007, Pittsburgh: Letang was ranked as the 10th best prospect on the Pittsburgh Penguins in the 2006 FW, then after scoring 6 PTS in the 2007 WJC rose to 10th best prospect in the whole NHL. He wasn't just on the gold medal winning team, he was also the captain.

3) Thomas Chabot Canada 2017, Ottawa: Being named MVP of the WJC can really boost your stock. Chabot ranked as the 38th best NHL prospect in 2016 FW, then after a sensational WJC in 2017 with 10 PTS as MVP, he shot up to the 3rd spot in 2017 FW.

4) John Carlson, USA 2010, Washington: Carlson was the 27th ranked prospect in 2009 FW, then he scored 7 PTS in the WJC and boosted his rank to 5th in 2010. He finished that season playing in the NHL, and never looked back.

5) Christain Dvorak, USA 2016, Phoenix: Dvorak ranked as the 48th best NHL prospect in the 2015 FW and jumped up to the 10th spot in 2016 after 8 PTS in the WJC.

6) Partick Berglund, Sweden 2008, St Louis: Berglund ranked as the 57th best NHL prospect in 2007 FW, then climbed up to 5th in 2008 after 7 PTS in the WJC.

7) Anthony Duclair, Canada 2015, Rangers: Duclair only ranked as the Rangers 6th best prospect in 2014 FW, then jumped all the way up to 17th in the whole NHL in 2015 FW after a great showing for Canada at the WJC.

8) PK Subban Canada 2009, Montreal: PK fell to the 2^nd round of the draft, and it became evident quickly that he had fallen too far. Less than a year after falling in the draft, he ranked as the 69th best NHL prospect in 2008 FW, then after scoring 9 PTS in the 2009 WJC, he was promoted to the 15th.

9) Derek Stepan, USA 2010, Rangers: Stepan was the Rangers 5th best prospect in 2009, then after 15 PTS at the 2010 WJC, he shot up to the 16th best prospect in the whole NHL. He was captain of the American team that upset Canada.

10) Joel Eriksson Ek, Sweden 2017, Minnesota: He ranked 23rd in the 2016 FW and climbed all the way up to 5th in 2017 after a strong 9 PTS at the WJC.

11) Maxim Goncharov, Russia 2009, Phoenix: He was a 5th round pick in 2007 who did not get ranked in the 2008 FW. He scored 6 PTS for Russia at the 2009 WJC and was ranked as the 21st best NHL FW prospect shortly after. This did not translate into NHL stardom.

12) Jamie Benn, Canada 2009, Dallas: A 5th round pick in 2007, Benn was already FW relevant by 2008 ranking as the 3rd best prospect in Dallas. After scoring 6 PTS at the 2009 WJC, he climbed up to the 33rd best prospect in the NHL.

13) Mark Stone, Canada 2012, Ottawa: He was a 6th round pick who did not get ranked in the 2011 FW. Then after leading Team Canada in scoring with 10 PTS in the 2012 WJC, he became the 45th best prospect in the NHL.

14) Johnny Gaudreau, USA 2013, Calgary: Johnny H ranked as Calgary's 4th best prospect in 2012. In the 2013 WJC he led the tournament in Goals with 7 and helped USA win gold. He was the 50th best NHL prospect shortly after.

15) Filip Forsberg, Sweden 2014, Washington: He ranked as the 21st best NHL prospect in 2013 FW, then after scoring 12 PTS at the WJC jumped up to 5th. Shortly after being ranked as the 5th best prospect in Future Watch, he was traded to Nashville for Martin Erat, in one of the all-time terrible trades.

16) Mikael Granlund, Finland 2012, Minnesota: Scouts already knew Granlund was really good when he ranked as the 20th best NHL prospect in FW 2011. Then after scoring 11 PTS in the 2012 WJC, he became the 2nd best NHL prospect in FW.

17) Jordan Greenway, USA 2017, Minnesota: He burst onto the scene with a strong performance in the 2017 WJC, with 8 PTS and dominating physical play. He had been Minnesota's 5th best prospect in 2016, then jumped up to 36th best NHL prospect a year later. 

18) Max Domi, Canada 2015, Phoenix: Domi was not selected to team Canada in 2014 but still managed to rank 18th in that year's FW. After a very strong 10 PTS in 2015, he was promoted to the 5th best NHL prospect.

19) Brad Marchand, Canada 2008, Boston: Marchand did not get ranked in the 2007 FW after being drafted in 2006. He wasn't even listed as one of Boston's 10 best prospects. Then after scoring 6 PTS at the 2008 WJC, he became the 64th best NHL prospect.

20) Anthony Cirelli, Canada 2017, Tampa: Cirelli went unranked in the 2016 FW, then after a successful WJC with 7 PTS, made it as the 72nd best NHL prospect in 2017.