Did Chris Ballard Make a Good Trade?

Last night, I slogged through 3 hours of NFL bloviating only to be rewarded by Chris Ballard strapping my patience to the rack and turning the crank to the “not tonight, dear” setting. Seriously, you can really damage a guy that way.

In the 2019 NFL draft, the Colts traded away their 1st round pick to Washington for two 2nd rounders, one of which is payable next year. So, the immediate question is, was it a good trade? I don’t pretend to know anything about college football scouting. I’ve never even done a mock draft, but that’s probably because I have no access to any published collection of information, like some kind of a guide for the draft.

Anyway, my point is that I have no idea how to judge a draft trade from a “team needs” perspective and I won’t even try. However, as a numbers guy, I can quantify the hell out of it and provide some context to the value given up and received in the deal.

---

JIMMY JOHNSON

I’ll start with the trade value chart made famous by former Dallas Cowboys coach Jimmy Johnson. For those unfamiliar, Johnson created a table that assigned a value for each draft pick position and since earlier picks are far more valuable than later picks the values had a steep initial decline.

I normalized the chart to decline from 100% so you can’t see the actual values, but according to this math, last night’s trade looks like this:

The exact position of next year’s 2nd rounder is unknown, so I just used #46 as an estimate, which gives the overall trade a surplus of 180 points. That surplus is equivalent to a free mid-3rd round pick (#82). Woohoo!

Of course, the 2020 draft is a year away and so that 2nd round pick should be discounted. I have seen suggestions of a full round discount which would make it a #78 pick worth only 200 points turning the 180 surplus into 60 point deficit. Boo!

I think that a full round discount is far too high, but you get the idea of the uncertainty and sensitivity around the numbers. With such a steep decline, small differences can have large impacts.

---

HISTORICAL TRADES

In Jimmy Johnson’s world, a #1 overall pick is worth five #31 picks and because that curve is so steep, it is been discounted by armchair GMs everywhere, who assume that no one uses it (and they are probably right).

Rich Hill — managing editor of Pats Pulpit and all around stats guy — put together a chart based on the actual draft trades that GMs have performed. According to his data, this is the actual trade value curve:

It’s actually a bit steeper in the first round than the JJ curve, which means that GMs are valuing early 1st round picks even higher. An overall #1 pick on this curve is worth almost 5-1/2 #31 picks. That may seem like a steep price but keep in mind this is based on actual historical trades.

According to Rich’s math (which uses a different scale), the Ballard trade is:

Without discounting the 2020 pick, the trade provides a surplus equivalent to an early 4th rounder (#102) which I guess could be wiped out if you discount heavily (which I don’t).

---

MASSEY AND THALER AND STUART (oh my)

Wharton professor Cade Massey and Nobel Prize winning Economist Richard Thaler wrote a couple of papers that compared the market value of draft picks with the historical value of drafted players. Their findings were that, in general, GMs overpay to trade up into the first round. They utilized complex analysis using compensation and player performance to measure the value of each pick, which I certainly can’t recreate (let alone fully understand).

However, I will leverage (steal) independent work done by Chase Stuart of Football Perspective to give you a simple analogy of what they did. Chase put together a draft value chart based on the career AV of the drafted player by pick position.

As you can see the curve of actual player performance is not nearly as steep as the other curves. In the AV world, a #1 pick is worth only 2.7 #31 picks, almost half of Rich Hill’s actual trade curve. This means GMs value early round picks too much relative to what they get out of the players. It is partially this disparity in performance vs price that led Massey-Thaler to their conclusions.

Now, I know a lot of you don’t like AV, but even if you use a different production measure like games started or games played or even snap counts, you will see very similar, shallower curves and they all show that it is probably better to trade out of the early first round as GMs are willing to give up more than they actually get.

According to Chase’s model, Ballard’s trade looks like this:

That surplus is worth a free #82 pick which is exactly the same as the Jimmy Johnson trade value. Hey! what gives? Shouldn’t the AV curve reward trading out of the first more than the other, much steeper curves?

Unfortunately, the #26 pick was already too far down the curve. Basically, the slope between #26 and #46 on the JJ curve flattens to the point that it looks similar to the Chase AV curve. Crap, if only we had lost more games last year!

---

CONCLUSION

So overall, this trade looks good for the Colts, but it doesn’t look great like last year’s absolutely-crushed-it-and-took-them-to-the-cleaners trade with the Jets:

Here are all of last night’s trades listed from the perspective of the team trading down. I think relative to the other trades, Ballard did well.