NHL Entry Draft: The costs of moving up

A hot start to the season had the Montreal Canadiens seemingly on track for a deep playoff run. Effective forward lines, depth (perhaps too much) on defence, and the reigning Most Valuable Player in net backing them all up propelled them to a franchise-best start.

Less than two months later, with their star goaltender on the shelf, things began to unravel, and the team plummeted down the standings, going from being first in the entire league on December 3, to 22nd once the final regular-season games had been played.

While that did make for a disappointing season and an abrupt end to what had been a promising campaign, it left Montreal in the ninth-overall slot of the entry draft. A chance at winning one of the top three picks came with that disqualification from the playoffs, but luck was not on the team’s side, and they held onto their original position.

That pick should net a decent prospect for the organization, one who has a legitimate shot at playing in the NHL. But by that point in the draft, the probability of drafting a star prospect is lower than it is for one of the very top picks, with surefire NHL stars usually selected among the top few slots.

After a few years of trying to fill roster needs with free agents who have rarely panned out (for whatever reason in a particular case), Marc Bergevin, his management and scouting staffs may decide that it’s time to make a deal to move into one of the spots near the start of the draft, and add a potential star to the system.

Bergevin has mentioned his desire to make such a move in the past, though has only been able to pull it off in the virtual world to date. The cost of finalizing such a move has likely been seen as too steep for the Canadiens’ General Manager, but a failed year in what was supposed to be a season for his team to contend for a Stanley Cup may have him entertaining the idea of creating a more significant offer.

Moving up

The current administration’s most recent trade-up in the draft came in 2014 when they packaged the club’s third- and fourth-round picks (87 and 117, respectively) to move up to the 73rd spot to select Brett Lernout. In 2010, the organization made its last transaction to move up in the draft’s opening round, combining their low first- and second-round selections in exchange for a fourth-round pick and a five-spot jump — from 27 to 22 — choosing Jarred Tinordi.

Teams are reluctant to part with top picks. Three teams moved up in last year’s draft, but the highest pick involved in those trades was 24th overall (going to Philadelphia). In fact, the last top-five pick to switch hands came in the trade between Boston and Toronto that saw Phil Kessel donning the Maple Leaf, and that change only happening after the Leafs finished as one of the bottom teams in the league the season after the trade was made.

The perceived value of picks seems to fall quickly in the first round, to the point where general managers believe there are few packages to make giving up their top spots for several lesser assets a worthwhile tactic.

The worth of a pick

Eric Tulsky looked at the trades involving just draft picks from 2006 to 2012 to understand the value management staffs across the league place on each of the top 100 positions in any given draft year.

The result confirmed that widely-held belief: picks near the start of the draft are thought to have a value much greater than those later in the event, with a first-overall pick being worth substantially more than even the second selection.

Perceived value of the first 100 picks, based on NHL trades from 2006-2012

According to the analysis, even the fifth-overall pick is seen as only half as valuable as the first, with the number falling to around 20 percent by around the 20th draft choice. Anything beyond a second-round slot (after #60) was basically regarded as a throwaway worth very little in return.

Attempts have been made in recent years to quantify the actual value of a draft pick using various metrics. Colorado Avalanche blog JibbleScribbets performed an analysis by looking at average time on ice for players selected in certain ranges of the draft. Michael Schuckers of Statistical Sports Consulting used NHL games played to find the worth of each of the first 210 picks in drafts from 1988 to 1997.

Scott Cullen assigned a number from 1 to 10, with 10 being a career of 10 or fewer NHL games, and 1 being a player judged to be a generational talent that has become a superstar on his way to the hall of fame, to each prospect. The result was an average quality of player selected at each of the top 30 positions in drafts from 1990 to 2009, with picks in later rounds being combined into groups.

For Sportsnet, Stephen Burtch used a two-pronged approach, looking at the probabilty of a prospect playing a certain number of games (60, 100, and 200) at each draft position, then multiplying that by the average offensive production for each pick. The final result suggests a similar trend to what GMs perceive about pick value, with the fifth-overall selection about half as effective as a player chosen with the first pick. Looking at just the games played aspect, however, you get to fifteenth overall before the value drops to 50% of the top pick.

Taking all of these (and a few more) methodologies and combining them into a single valuation, hockey analytics writer PDwhoa recently published the results of that project on Puckalytics.com.

The aggregation shows that the notion of top-three picks being highly valuable matches the reality, with quality dropping quite dramatically afterward. The difference between the top pick and the second isn’t nearly as pronounced as it is in the first chart, but it is still the second-largest gap in value between consecutive picks, behind only the massive gap from third to fourth.

That difference legitimizes the reluctance to part with a top-three pick, but a relatively gradual drop in value — both perceived and actual — over the remainder of the first round suggests that trades of picks four through 30 should be entertained more often.

Making a trade

In Montreal’s case, their pick at ninth overall is worth 1.63% of the total draft value according to PDwhoa’s consolidation, while pick number four is worth 2.02%. In theory the Canadiens could combine their first-round pick with one worth about 0.40% (their third-round pick at 69th has an actual value of 0.44%) and offer a team relatively equal value for the right to draft the next player after the big three are taken off the board on June 24th.

Looking at the perceived value graphic, however, there are about 14 points separating fourth and ninth. To make up that difference, a team would have to offer a pick around the end of the first/start of the second round to compensate, making for a steep price. Using the actual value numbers, that would require Montreal’s first- and second-round picks, worth a total of (1.63 + 0.67 =) 2.30% of the draft’s available value to acquire a pick that has been shown to be worth less than that amount.

Ironically, because of the discrepancy in how much picks are thought to be worth, and how much value the analyses suggest they actually have, a team would be better off trading up with a stats-savvy club who had a more concrete understanding of the value of a given pick.

It may come to be the case that a team regards a certain player in much higher esteem than anyone slated to go before its next selection, and in that case it may be beneficial to ignore the average value of the picks possessed and make a move to claim the desired player. That could mean overspending, like the scenario outlined above, or perhaps even adding a redundant prospect to the offer to move up a few spots and claim a talent you’re really high on.

If that’s the game-plan, you’d need to be confident that the player will become a key contributor to your team, because scouting isn’t an exact science, and there is plenty of available value in those later-round picks.

Which player should Montreal trade up to select in the first round (ranking in parentheses)