2026 NHL Draft

[dbr]

I see the discussion has already touched on this, I got a bit bored earlier and thought I'd pull some data from 20 drafts (2000-2019) to see if we can quantify pick value a bit more accurately - just the high level hockeydb stuff along with some TOI/G career averages to help assess defensemen.

(I left goalies out because they are fucking voodoo)

Did some assessment of guys careers based on career games played and then tried to refine that a bit further (albeit still very shallow analysis) using career averages of points per 82 for forwards and TOI/G for defenseman - a "star" bucket for 70+ or 22 minutes+, a "core player" bucket for 60+ or 19 minutes+. Any draft pick with 400+ GP is a success, any pick with 200 or more was still a player for some stretch, anything else is a failure.

For the entire set, 4163 non-goalies drafted, 75% of picks bust completely. Just 8% are "core players" and just 2% go on to become "stars" by the definitions above.

Code: Select all

Pick Number	Picks		Bust Rate		Core Player%	Star Player %
1-5		96		5%			64%			31%
6-10		96		15%			29%			18%
11-20		192		27%			29%			8%
21-30		187		43%			12%			2%
31-40		177		64%			10%			2%
41-50		186		66%			9%			1%
51-60		184		64%			8%			2%
61-70		180		76%			8%			2%
71-80		175		78%			4%			2%
81-90		172		84%			3%			1%
91-100		180		81%			3%			1%
101-133		590		84%			5%			1%
134-167		583		89%			3%			0%
168+		1165		91%			3%			0%

Obviously there are a few significant drop-offs here but to me the practical one (given how hard it is to get top 10 picks) is at the end of the 2nd round.

[Picker of Cherries]

Verhoeff, Willander, Hronek all have the potential to be top pairing RHD, which is probably overkill. Selecting Verhoeff opens up the option to trade one in a few years. If Hronek consistently plays the way he has this season, the return should be quite large when his NMC expires in three years. Lots can happen, but selecting BPA, is always the best strategy.

[Cornuck]

Nice work, dbr.

Top 5 seems to make a difference.

[Meds]

Code: Select all

Pick Number	Picks		Bust Rate		Core Player%	Star Player %
1-5		96		5%			64%			31%
6-10		96		15%			29%			18%
11-20		192		27%			29%			8%
21-30		187		43%			12%			2%
31-40		177		64%			10%			2%
41-50		186		66%			9%			1%
51-60		184		64%			8%			2%
61-70		180		76%			8%			2%
71-80		175		78%			4%			2%
81-90		172		84%			3%			1%
91-100		180		81%			3%			1%
101-133		590		84%			5%			1%
134-167		583		89%			3%			0%
168+		1165		91%			3%			0%

Obviously there are a few significant drop-offs here but to me the practical one (given how hard it is to get top 10 picks) is at the end of the 2nd round.

Unless I’m misunderstanding what you mean or I’m misreading the numbers, it appears to me that the most significant drop offs are after the 20th (16% increase in bust) pick and then after pick 30 (19% increase in bust).

The difference between the late 2nd round and 3rd round is only 12%.

[dbr]

A couple of things, one is that I was focused on the rightmost columns (percentages of picks that turn into impact players) more so than avoiding busts.

You’re right about where the big numerical drops are, 20-30 is significantly worse than 11-20 for eg. - but it’s hard to plan for having a pick in one range or another, whereas “draft outcomes after the second round decline to the point where they’re not all that different from one another” might mean you look to trade up into the 2nd more, or prioritize those picks in trades on what have you.

[Meds]

Oh fuck me. I’m on my phone and didn’t realize that there more columns I could scroll over to. I only saw the first 3 columns.

Disregard.

[2Fingers]

Oh fuck me. I’m on my phone and didn’t realize that there more columns I could scroll over to. I only saw the first 3 columns.

Disregard.

Ha ha me 2, was going to ask for more data but I see it is provided.

[Aaronp18]

Well we clearly need an Edmonton like run of top 5 picks here just to be safe!

[Ronning's Ghost]

I got a bit bored earlier and thought I'd pull some data from 20 drafts (2000-2019) to see if we can quantify pick value a bit more accurately - just the high level hockeydb stuff along with some TOI/G career averages to help assess defensemen.

The analysis I should have done, but was neither sufficiently bored nor sufficiently ambitious.

(I left goalies out because they are fucking voodoo)

Most draft analyses I have seen take the the same approach -- right down to the terminology.

But still, I think more of the better goalies are drafted earlier than later.

[dbr]

Most draft analyses I have seen take the the same approach -- right down to the terminology.

Yeah, I’ve read a few and knew I should take the copout

But still, I think more of the better goalies are drafted earlier than later.

You could probably analyze this, but you’d have to look at them more on a case by case basis. The games played threshold has to be lowered so you get guys who never really established themselves alongside Roman Cechmanek

[Topper]

Have no fear dbr, no one will remember it in a week as they rave about the mid third round pick acquired for Garland

[Madcombinepilot]

Code: Select all

Pick Number	Picks		Bust Rate		Core Player%	Star Player %
1-5		96		5%			64%			31%
6-10		96		15%			29%			18%
11-20		192		27%			29%			8%
21-30		187		43%			12%			2%
31-40		177		64%			10%			2%
41-50		186		66%			9%			1%
51-60		184		64%			8%			2%
61-70		180		76%			8%			2%
71-80		175		78%			4%			2%
81-90		172		84%			3%			1%
91-100		180		81%			3%			1%
101-133		590		84%			5%			1%
134-167		583		89%			3%			0%
168+		1165		91%			3%			0%

Nice work, man!