My NHL Draft Manifesto

June is draft month and we celebrate that hard here at PPP. I am very lucky in that I get to read all of Brian's draft posts early, and it becomes more like a non-fiction narrative about the coming draft, told in a unique way. I love reading these pieces, and I always end up with a favourite who someone else drafts.

I get a lot out of talk about draft prospects and I genuinely love the Top 25, even in this jaded age where the done thing is to notice that the quality of the prospects isn't that great overall. The reason I love this stuff is because watching younger players in simpler leagues and then seeing where they end up gives a coherence to longer stretches of time than you get with NHL seasons that reset every summer to a new team with new faces. You also learn things if you keep your mind open.

I like Brian's articles particularly because of how he thinks about prospects in terms of skills that can show improvement and those less likely to. Over the years, he's developed ideas about defencemen and what they need to be able to do that match my own very closely, and it's funny that we never talked about it, he just wrote something, and I wrote something and they more or less said the same thing.

I think these ideas about defence are shared by Brad Treliving and Craig Berube to a very large degree, and not by Kyle Dubas or Sheldon Keefe. And this is the area where I think Treliving has succeeded the most in making the Leafs into a different team.

The other thing Treliving has done is traded draft picks in a way I strongly disapprove of. He's kept far too many.

The truism that teams have to draft and develop is not. True that is. You can use some picks, and when you get a hit, you should cherish that player and trade them for someone you really need. Unless they are ready for the NHL and you actually really need someone just like them, but that is rare if you're not drafting in the top 10. This calculation is often phrased as wins now vs wins later because people are constantly trying to find the thing that functions inside the NHL like money does outside it. And they will never find it. More on that grumpy topic later.

But this is supposed to be a manifesto of my beliefs, so a list of them is in order.

Draft round is irrelevant

This is true: every third person who has R Studio or Tableau or knows Python has made a draft pick value chart in the last 20 years. There are thousands of them, and they all look the same.

Only one person who has drawn one has ever gone on to draft someone. So far!

(This is a major part of my draft season fun – making Tulsky jokes while using his ancient work at Broad Street Hockey.)

Okay, the main point of the draft is broadly the top 10-15, and what that second number lands in a given year is where "deep" or "not deep" draft characterizations really reside. Second is a chunk of picks that end around 25 to 30. The third tier is everything from 25 or so to about 45, and the fourth tier is all the rest which is an undifferentiated mass.

So here are the things that aren't true:

1. Draft round number tells you where in the lineup the player should land.
2. A third rounder is way better than a seventh.
3. Most first-round picks become career NHLers.

And the things that are true:

1. The vast majority of the junk picks in the fourth group get you nothing.
2. Second round picks are devalued by many GMs relative to low first-rounders in trades. Trade the first, keep the second.
3. Goalies are the exception to everything, and outside the top 10, draft pick number has no relationship to goalie success in any way.
4. Bonus true thing: all draft pick charts that use NHL games played to set the value use such a low bar for NHL success, they grossly exaggerate the ultimate value. Also known as the Freddie Gauthier rule: I don't care that the Leafs drafted him, that's not a hit.

Almost all the late round surprise hits in the recent NHL past were drafted late for very good reasons. Predicting the future of teenagers is not an exact science, but the NHL does very well and there is some evidence they've been getting better in the last decades.

Probability of success is very low

There's 200+ players drafted every year and there are usually less than 50 rookies who become fulltime NHLers each season. Almost all of the successful picks are in the top 15.

Mining for rare gems requires luck and one other thing

You can have the best scouts in the world and still miss out on players who turn out to be great deep in the draft. For the impossible to guess at big hits on bad picks, you need luck.

The only sure way to get more luck is to have more picks. In other words, trading a couple of picks at the deadline isn't going to kill your draft day. You have to have more than one extra loony to shovel into the slot machine. Draft day is go big or stay home and read the scouting reports on actual players.

Draft pick value, like player value, is not immutable

This is one of my core beliefs that goes well beyond the draft. It's true for all asset-value interpretations of the NHL. Value of a pick, a player, a prospect, their impact on game outcomes, their future worth – everything – is contextually shaded and varies from team to team and situation to situation.

A defenceman is worth more to the team that needs a defenceman. Another draft pick matters more to a rebuilding team with 10 or so already – they're bulk buying prospects. A trade of a pick for a rental – the despised deadline deals – are not trades of wins in the future for wins now, they are trades of a pick that could potentially turn into a player who might play NHL games at some time vs one who absolutely will right now. When it matters.

There is no "market" and no market value and applying stock-trading logic like buying low and selling high (I've done this myself, but I should stop.) misses the rather key point that there is no money here. In stock trading, money is the entire point so you can't make the NHL analogous. You aren't buying wins or goals in a trade or at the draft, and you aren't paying for them with wins or goals because you can't assign a win or goals value to a player or a pick that will always be true.

I betcha E. Tulsky tries. But even he has to concede that the value of a forward who can actually score goals when your team continually scores under expected is worth more than another puck mover with a higher theoretical value. At least, he should.

The Leafs have too many picks

Hockey is a conservative world where every action is subject to review in hindsight on how it turned out, not what your reasonable expectation of the probabilities would be when you made your decision. Except in Carolina and likely Pittsburgh, maybe Long Island now.

Given that, GMs hesitate to be too bold, too radical. Offer not valid in Vegas and Tampa Bay. Did Brad Treliving Trade Away the Future of the Leafs is such an easy headline, even he hesitates to trade enough picks.

Picks are junk. That's my real feeling. Are you in the draft lottery? Nope? Trade them all, keeping back the occasional second to actually use.

The Leafs have one second rounder – good, and five junk picks – bad. All of them should be traded.

But prospects, though!!!

Let's talk about that in August when everyone is shocked and amazed at how bad the Leafs' prospect pool is. If only they had had some more extremely low probability prospects taken with junk picks over the last few years, there'd be more not very good prospects upping the number of arguments, but not the value of the group.

And yes, with apologies to Brian and his genuine foresight, impressive scouting ability and fannish love, trading Greb was the right thing to do.

Matt Knies is not a repeatable skill

That's it, that's the whole point. You can't set out to get a guy who is that good and will hit the NHL that early with a pick outside the top 10-15. I mean, not unless you shake the pick in your hand first and blow on it for luck.