We’re not totally sure the offseason is over, but given that real games are about to start, it’s probably best to consider it so. With that, it’s time for our semi-annual look at how our model performed against the trade market.

As a reminder, we keep a log of every trade that occurs in real life – big and small – as a check on our model. This allows us to see patterns, and to make adjustments if necessary. Since we’ve been doing this for almost five years now, at this point we feel like any adjustments will be minor. Here’s how we did:

 

2023/24 Offseason results:

• Number of real offseason trades: 70
• Number of trades accepted by our model: 65
• Hit rate: 92.9%
• Average variance: 3.1

 

Cumulative results, August 2019-Present:

• Number of real trades: 641
• Number of trades accepted by our model: 602
• Hit rate: 93.9%
• Average variance: 2.0

 

As you can see, this offseason’s results were very much in line with our history, which suggests our model is relatively stable. The variance went up a little, but the hit rate was very close to our 5-year average.

But, to be fair, we’re counting all the little ones (including all the marginal players for, say, cash considerations), which are fairly easy hits, as well as the big ones, which are more challenging. 

So what if we just looked at the big ones? By this, we mean the ones that Jeff Passan or Ken Rosenthal announce. These started with the deal that sent Mark Canha to the Tigers in early November, and continued through the trade of Dylan Cease to the Padres in March. 

 

2023/24 Offseason, major trades only:

• Number of real offseason major trades: 33
• Number of trades accepted by our model: 29
• Hit rate: 88%
• Average variance: 4.5

 

Compared to the overall results of the offseason, the hit rate went down a little, and the variance went up a little – which is historically normal.

Now let’s break down those 33 major trades further, into three categories: the hits, semi-hits, and misses.

• Hits: 16
• Average variance: 1.9

Going chronologically, there were 16 trades that were close enough to be accepted by our model without equivocation: Mark Canha to the Tigers, Cal Quantrill to the Rockies, Luis Urias to the Mariners, Vidal Brujan to the Marlins, Jarred Kelenic et al to the Braves, Alex Verdugo to the Yankees, Marco Gonzales to the Pirates, Tyler O’Neill to the Red Sox, Evan White to the Angels, Ray Kerr to the Braves, Estevan Florial to the Guardians, Chris Sale to the Braves, Corbin Burnes to the Orioles, Ross Stripling to the A’s, Gregory Santos to the Mariners, and Dylan Cease to the Padres.

 

• Semi-hits: 13
• Average variance: 5.1

There were 13 trades that were a bit off, as either overpays or underpays, but still accepted by our model: Aaron Bummer to the Braves, Scott Barlow to the Guardians, Kyle Wright to the Royals, Eugenio Suarez to the Diamondbacks, Tyler Glasnow to the Dodgers, Adrian Houser to the Mets, Luke Raley to the Mariners, Andrew Kittredge to the Cardinals, Cristian Mena to the Diamondbacks, Caleb Ferguson to the Yankees, Steven Okert to the Twins, John Schreiber to the Royals, and Manuel Margot to the Twins.

 

• Misses: 4
• Average variance: 11.8

Finally, there were four trades that were rejected by our model: Juan Soto to the Yankees, Robbie Ray to the Giants, Michael Busch to the Cubs, and Jorge Polanco to the Mariners.

The four misses were quite prominent – the Soto trade was the biggest move of the offseason, save for the Ohtani signing. 

Looking back at that one, we felt justified in our surplus value number of 23.8, considering Soto is set to earn $31M in salary in his final year of control, which, added together, implies a market value of $54.8M. 

No player in history has ever come close to that in AAV, so even if you add in a superstar premium, a bonus for possible postseason contributions, and the value of a draft pick after a presumed rejected QO, a $23.8M premium is still a hefty estimate on top of the $31M you already have to pay him. 

Yet the Yankees blew well past that number in their package, coughing up $57M in surplus to San Diego (but also acquiring Trent Grisham, which closed the gap some), proving how badly they wanted Soto, after a disappointing 2023 season. So yeah, we can live with that miss.

The Robbie Ray/Mitch Haniger trade was a swap of underwater contracts that was heavily dependent on injury-risk estimates, since both Ray and Haniger have serious issues in that department. Chalk that one up to unknown factors – we don’t have access to the medicals, so if we’re a bit off on the health estimate, that’s why.

The Busch trade to Chicago was due to a roster crunch, as the Dodgers needed to clear a spot on their 40-man roster for Teoscar Hernandez, who was signed the next day.

As for Polanco, he had been heavily coveted by the Mariners for years, reportedly, and Jerry Dipoto has no trouble overpaying for guys he likes (see: Luis Castillo).

 

Summary 

We were particularly pleased that we hit on two major starters – Corbin Burnes and Dylan Cease. As with last year’s deadline deals of Max Scherzer and Justin Verlander, it suggests we’re doing quite well with our values on big-name starting pitchers in general. And interestingly, virtually all of Alex Anthopolous’ flurry of deals came in as fair or very close to fair in our model – once again validating the model’s framework.

As a reminder, our model is based on the concept of surplus value, which is the amount of a player’s field value over (or in some cases, under) his salary. Because front offices have to manage to budget limitations set by ownership, the money matters. Anthopolous’ deals all reflected the importance of salary in the overall picture.

That said, sometimes teams are desperate to win, and will over amp on field value. You want Soto? Pay up.

And while we’d love to see the variance number come down a little, overall, we’re content with these results – and we’ll continue to track the market as best we can.