Evaluating the manager of an MLB team is a difficult task. Of course watching the team every day, and seeing how the manager handles various situations we can get a sense or feeling of how he is doing. But it’s almost impossible for us to separate our bias, whether it’s bias towards the individual or bias towards how we believe the game should be played and managed.
There are also many situations where we just don’t have enough visibility into the information necessary to judge the manager’s actions and the supposed advantages or disadvantages he has going for him.
Beyond our feelings and bias, we, like upper management often turn to blunt instruments. That may include simply looking at wins and a losses. Or perhaps attempting to judge manager performance by an underlying peripheral, such as Pythagorean Won Loss. The issue with that approach however is that any gap between “pythag” and actual WL is extremely dependent on a teams record in one run games. And the one run games are most often determined by some combination of reliever performance, “clutch” or timely hitting, and luck. It can be extremely difficult to tease out the primary factors influencing a team’s record in one run games in any given year, let alone figure out the manager’s role in those factors.
While thinking about these factors I stumbled across an approach that started to pass the “smell test” , at least for me. (Bias alert) Instead of looking at pythag WL vs actual, how about expected WL from WAR vs Actual WL ? WAR is context neutral. It’s measuring the overall hitting, pitching, and fielding performance without taking into account the inning or leverage. This gives us a base “talent performance” to start with and does not muddy the water with vagaries of the luck of the bounce, or the umpire’s call that can and often does have an outsized effect on the outcomes.
So what does that look like since 2017 ? How have Torey’s teams’ results differed from the base talent available to him ? I’ll start out with the first most important table. This table takes the average WAR from both Baseball Reference and Fangraphs, then adds that total to the average replacement level win total. WAR is based on a team replacement level Win % of .294 , or 47.6 wins in a 162 game season. You simply take the player’s aggregate WAR, add it to the replacement wins, and you come up with an expected wins. (2020 is obviously prorated against 60 games).
Keep in mind the Batters numbers include fielding. So for example while the 2018 hitting was below NL average by a fair bit, (88 wRC+ vs. NL avg 94 wRC+) their fielding was great, and they picked up a lot of WAR through defensive runs.
What the above table is telling me is that Torey’s teams have been pretty close to expected wins when compared against base performance, up through 2020. 2017 & 2020 are within rounding errors, and 2018-2019 team’s actually won a few more games then we might otherwise have expected. In total, 2017-2020, the team actually won about 7 more games than expected.
However 2021 was a disaster for Torey and coaching staff. They won 10 fewer games than they should have been expected to win based on the roster he had on hand. To be sure, the 61.5 expected wins is a bad total. Injury played a part, as did being forced to play certain guys out of position. But there was enough talent there to go 62-100 at the very least, instead of the 52-110 they ended up at.
Torey had a bad year. There were many things he didn’t get right. The way he platooned players and used guys in the field just did not work out. The pitching staff was first decimated by injury, but there were many seemingly questionable choices at various points in the season, more than usual. The 10-31 record in one run games didn’t happen ALL because of luck.
To be sure, lack of timely hitting played a role. The team’s wRC+ for the year of 84 was 11 points lower than the NL average, BUT in the 7th, 8th, 9th, & Extra innings in HIGH LEVERAGE they had a 68 wRC+ vs. league avg 91, a whopping 23 wRC+ points worse than the league, even when you control for shutdown late inning relievers. (Note the league avg is against the other 14 teams, with the Dbacks removed from the avg). What this means is as bad as the hitting was, it was much worse when it counted most. So it wasn’t only the bullpen.
Why didn’t Torey shake things up ? For example, he shut down the running game. While attempts had been slowly declining each year, they were at an all time low of just 59 last year.
For some strange reason, Torey stopped using the replay challenge as well. While it’s true that video review is done outside the dugout and the video review coordinator is responsible to tell Torey if he’s got a shot or not, this is just bizarre. How do you go from 30-40 reviews to just 10 ? and Go from getting half right to just TWO of them ??
Obviously we don’t want to overstate the importance of video challenges. But whether it was failure to shake things up on offense in the midst of long losing streaks, playing guys out of position, never challenging a call, or sticking with struggling pitcher too long, (or sticking to the pitch count too closely when a guy is cruising) something was clearly off.
In the four previous seasons it can be argued that Torey did what you would have expected with the talent he was given and how the players performed on the field. After all, he can’t get in the box or on the mound. Sometimes he managed to squeeze a few extra wins out of his team perhaps. But I wouldn’t exactly try to make the case that he was getting his teams to over perform either. For some reason, in 2021 he was like a deer caught in the headlights.
I think the team did the right thing in how they handled his contract. Hopefully it was just a bad year for him. We all have them.
It would be interesting to make similar comparisons for other managers around the league and also former D-backs managers. I might get to that at some point