Trading Up vs. Trading Down

By #jontent (@yakovmironov)

One of the most popular topics for the next month will be debating whether or not teams should be looking to trade up or down at the draft. Generally the math behind it points to a fairly strong conclusion, and that is the team willing to trade down generally wins the trade because of premium being placed on the higher pick.

Last year the Leafs gave up the 24th overall pick to Philadelphia so the Flyers could select Travis Konecny, a talented center who was inexplicably being looked over. The fact that Konecny was there when the Leafs were going to pick was an exciting moment for a lot of Toronto fans, and seeing the pick flipped for the 29th overall (later moved) and the 61st overall (Jeremy Bracco) was confusing since Konecny was already seen as being more valuable than the 24th overall pick.


Using the draft pick value table from DTMAH (Don’t Tell Me About Heart) the Leafs gave up a pick of 445 value and acquired a pick of 410 value and one of 255, making it look like a clear win, but when you look at where Konecny was supposed to go, you can already see the value beginning to decline.

Using Platinum Seat Ghost’s Aggregate Draft Rankings from 2015, which compared a number of popular draft rankings, we see that Konecny was 16th overall. A value of 530, which still gives the advantage to the Leafs. We also see from these rankings that there was significant variance on where Konecny was expected to go. If he went 33rd overall as one service ranked him as, he would have carried a value of 380, but counter to that, if he went 7th overall as another service predicted, he would have carried a value of 700, meaning the Leafs would have actually received less value in return.

Looking at the draft in hindsight using the pGPS comparable %, Ryan Biech of found that Konecny would be the 16th overall pick if the 2015 was redone today, establishing the value around 530. It’s also worth noting that Konecny suffered a shoulder injury which could effect his long term value, and while he might have been the better player in the deal, this kind of event only goes to show the gamble associated with making this kind of aggressive trade up deal.

This example is just a quick summary of how draft pick value can fit into what gets looked at heading into the draft, but as DTMAH notes in his article…

Reminder that these value’s are arbitrary numbers and should only be used to compare draft slots and not any players involved in a potential trade. This once again is an approximation many years of data and in no way a hard rule of how every pick should be valued.

That’s what has always been my philosophy relating to draft value as well. If you are trading up or down, what is the combination value of picks that you would need or be willing to give up to do so. You know the value you associate with the player at the top of list, and what you need to make it work.

It does however need some consideration for what the actual draft looks like though. Looking at the first round there appear to be seven tiers of talent available based on the consensus of draft ranking services. There are the top 3, 4-8, 9-12, 13-14, 15-21, 22-23, and the rest. The interesting thing about the rest is that it basically continues out to pick 47, as that is the total number of players who are presently receiving some consideration as first round picks based on our consolidated draft rankings.


What we can see from this is that there are a few points where draft pick value and consensus ranking don’t necessarily correspond.

  • After Michael McLeod (12th pick) there is a steep drop off in the consolidated ranking score, but very little in draft pick value. You could easily make the argument that Brown, McLeod, and Bean are in the same class anyway so it’s not a big deal.
  • The slight drop off in draft pick value between 14 (Bean) and 15 (Fabbro) but the steep drop in consolidated ranking score is similar. Fabbro could very well be in that upper class, so there’s probably not a lot to it at this point. These are all good players and the draft pick value is reflected the benefit of having a bigger pool to choose from.
  • After Max Jones at 21 it starts to be a bit more of an issue. Jones is the last of the players viewed as locks to go in the first round, and that’s something that factors heavily into the consolidated rankings score, but there is only a 5 point drop in draft pick value between Jones at 21 and DeBrincat at 22. More importantly there is only a ten point drop between Tufte at 23 and Asplund at 24, and 24-47 in the Consolidated Rankings are incredibly close in score. Pick 47 has a draft pick value of 307.5

What I hope I’m illustrating with this is that there is more likely to be a higher value associated with picks where one of the consensus top 21 players are still available. If Jones or Kunin or whomever slide to around 25, it seems reasonable to still consider reimbursing teams like you are acquiring a mid-first rounder instead of a late one.

Back to whether it makes sense to trade up or down…

There’s no arguing that numbers generally support the notion that you’re better off trading down. It’s the logical move that gives you a greater number of lottery tickets while generally doing little to diminish the odds of success of your original ticket.

Looking at what the Leafs did by giving up the chance to draft Konecny, they acquired Travis Dermott, who went on to have one of the best seasons by a defenseman in the OHL this year. They picked up Jeremy Bracco, who was pegged by many to go in the first round last year, and was one of the USNTDP’s best scorers in their history. And they picked up Martins Dzierkals, who has had a very good season in the QMJHL, and is currently competing in the Memorial Cup. Not a bad haul for giving up a player who might see his career slowed by an unfortunate injury. Hedging your bets seems pretty smart.

Of course there’s something to going after a player who you believe a step above the rest, and that’s still where draft pick value becomes a handy reference point. Using the Leafs as my favourite example, we’ll assume that Pittsburgh wins the cup and Toronto owns the 30th and 31st overall pick. Two nice pieces valued together at 795 or 400 and 395 separately. In theory, this should be enough for the 5th overall pick, in reality, you aren’t getting anywhere close to that. What it does beg the question of, is whether it’s worth the overpayment to get into the middle of first round to land another blue chip prospect instead of having two very good chances at swinging for the fences at end day one and start day two.

Closing the gap with roster players is probably the best option. The value assigned to players at the draft is a strange thing, where a 23 year old defenseman, who can play in the top four of a bad team or top six of a good one holds the same value as a fourth round pick and a 21 year old struggling make the NHL is valued as a mid first round pick with a high second round kicker. Knowing that options like these are out there (if you deal exclusively with Edmonton) is a great argument for turning over your roster as much as possible on these two days, moving up and down as needed.

What Can We Possibly Hope To Learn From This?
  • The value of moving up using picks is incredibly steep, and you have to be damn sure about the player you want if you’re going to do it because initially it’s going to be perceived as an overpayment.
  • Trading down, or Bellichicking your way through the draft makes a lot of sense, but I’d be scared as hell of attempting this without trusting the guy who’s running your draft table.
  • If you are particularly confident in your trading ability, this is the time to show it off and you have a draft table next to Peter Chiarelli you are set.
  • Drafting where you are supposed to is boring and no one wants to watch three hours of people thanking the Sabres for hosting the draft and selecting the highest ranked guy on Bob McKenzie’s list. Hopefully teams will do interesting things given it’s not as high end a draft as last year, and potential expansion next year puts both a premium on certain teams to win now and for others to stockpile prospects that won’t be eligible for an expansion draft.

In closing: 

Teams I like- be smart
Teams I don’t like- be interesting

References not previously cited:
NHL draft: What does it cost to trade up?
Expected Value of NHL Draft Picks


1 Comment

  1. I generally agree with the article – I think the trade-up/down model breaks the closer to the top of the 1st rd you get. If there is a guy you like in the 10-30 range, I have no problems with the trading. I think where the no trade-up rule is more applicable is in the 3-7 rd. Wider divergence of opinion and lower quality prospects.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s