In 1987, Nolan Ryan won two legs of the pitchers’ triple crown, as he led the NL in both ERA (2.76) and strikeouts (270) – which helped him to a 6.4 fWAR, the top mark in the league. Yet, Ryan posted a W-L record of 8-16, which seems unfathomable. But it’s just one of many examples of why it’s a bad idea to judge pitchers by their win total.
But while we shouldn’t judge individuals by wins, it’s exactly how we judge teams. Wins are, after all, a team stat. So, can we judge a group of players on a team by wins? Is it okay to judge how the starters (or relievers) do in this team stat? It’s probably still better to judge by another number but let’s move forward with this concept, because it’s something we can grasp pretty easily.
Here’s a list of the 12 playoff teams in 2023 and how many wins they recorded from both their starters and relievers:
| Team | SP | RP | Total |
|---|---|---|---|
| ARI | 46 | 38 | 84 |
| ATL | 66 | 38 | 104 |
| BAL | 57 | 44 | 101 |
| HOU | 61 | 29 | 90 |
| LAD | 62 | 31 | 93 |
| MIA | 39 | 45 | 84 |
| MIL | 54 | 38 | 92 |
| MIN | 49 | 38 | 87 |
| PHI | 53 | 37 | 90 |
| TBR | 55 | 44 | 99 |
| TEX | 59 | 31 | 90 |
| TOR | 56 | 33 | 89 |
| 657 | 446 | 1103 | |
| 54.75 | 37.16666667 | 91.91666667 |
We see that playoff teams last year averaged 55 wins from their starters and 37 from their relievers. But there is a fairly wide range, with the Marlins’ starters amassing 39 wins, while the Braves’ SP totaled 66 wins. The range is a little more compact with relievers but still not overly so, as the Astros had 29 wins from their bullpen, while the Marlins’ pen posted 45 wins. It just goes to show there’s multiple ways to divvy up the wins necessary to make the playoffs.
The 2023 Mets got a healthy number of wins from their SP, as that group notched 49 wins. But their relievers did not pull their fair share, as they managed just 26. And, again, while wins aren’t a great measure of the value for starters and relievers, they do a good enough job describing the 2023 Mets.
In something that likely means very little, yet still jumped out at me, five starters for the Mets last year had winning records. Joey Lucchesi, Tylor Megill, Max Scherzer, Kodai Senga and Justin Verlander went a combined 40-24. Scherzer and Verlander are now on other teams, while neither Lucchesi nor Megill are likely to begin the year as starters.
Can the Mets’ revamped rotation help the starters to 50 or more wins this season? Can the bullpen contribute more than the 26 wins of a season ago? The hope is that the Mets can pull out somewhere around 85 wins. And it’s a worthwhile thought exercise to imagine where those victories are going to come from.
Can Senga, Jose Quintana, Sean Manaea and Adrian Houser combine for 40 wins?
Can Luis Severino and the depth starters combine for a dozen more?
Can any reliever come close to the 14-win season of 1986 Roger McDowell?
Can any reliever top the five wins of last year’s leader, Grant Hartwig?
Four times in his career, Quintana has reached double digits in wins, while Manaea has done it three times, with Houser and Senga having both done it once. It doesn’t seem outlandish to forecast 40 wins from this quartet. Of course, that will require all of them to be healthy and notch at least 25 starts apiece.
And it doesn’t seem unrealistic to expect a dozen wins from the remaining starters, figuring they’ll get somewhere in the neighborhood of 50 starts.
But it’s next to impossible to have any faith in what the bullpen will achieve, because of both the fickle nature of relievers in general and the question marks around the Mets’ particular bullpen group. Instead, let’s hope they don’t blow too many leads, while also pitching well enough to vulture some wins when they enter in tie games and the offense delivers late.
As mentioned earlier, wins are a team stat, given to pitchers who may or may not deserve them, based on rules that likely are outdated for how the game is played in the 21st Century. But it’s easier, for me at least, to wrap my head around what it would mean if the Mets’ bullpen somehow got 40 wins, rather than if they delivered a 3.81 ERA in 588 IP. For what it’s worth, the 2023 Mets pen posted a 4.48 ERA in 559 IP.
*****
Here’s something that was considered and then abandoned – not sure why. But after TJ’s comment, it seemed worthwhile to do.
| Team | SP | RP | Total | SP% |
|---|---|---|---|---|
| ARI | 46 | 38 | 84 | 54.76% |
| ATL | 66 | 38 | 104 | 63.46% |
| BAL | 57 | 44 | 101 | 56.44% |
| HOU | 61 | 29 | 90 | 67.78% |
| LAD | 62 | 31 | 93 | 66.67% |
| MIA | 39 | 45 | 84 | 46.43% |
| MIL | 54 | 38 | 92 | 58.70% |
| MIN | 49 | 38 | 87 | 56.32% |
| PHI | 53 | 37 | 90 | 58.89% |
| TBR | 55 | 44 | 99 | 55.56% |
| TEX | 59 | 31 | 90 | 65.56% |
| TOR | 56 | 33 | 89 | 62.92% |
| Total | 657 | 446 | 1103 | 59.56% |
| Average | 54.75 | 37.16666667 | 91.91666667 | 59.56% |
Interesting take – would be good to see the losses cut this way as bullpens that lose games suggest their weakness or relative strength. I suspect that the Mets bullpen cost the starters significantly more losses than many teams. Hopefully that will change this year
Fun article. It feels like the bullpen either pitched poorly and gave up runs and lost or the team did not score enough runs late to pull out games.
It would be fun to look at the losses too.
Very interesting Brian; would have never thought the numbers were like this. Every once in a while, we get a pitcher close to 4 ERA, but still had a great record. Those things are arbitrary most of the time, unless a starter got smoked. The problem for me is two fold:
1. Starting pitchers average barely over five innings a start, yet they can’t get a win unless they pitch five innings whereas a reliever can get a win throwing one pitch, getting a screaming line out and then his team takes the lead.
2. Starters can work their butt off, pitch great and have the bullpen give up the lead, only to have the team come back and someone else gets the win.
Quite interesting…but…I’m not sure I see any meaning in the breakdown, at least without additional information. Basically, I’d be fine if the Met pen had 15 wins, so long as the starters have 75, and vice versa. I’d but curious (but too lazy) to see how a team’s offense related to the wins by starter vs reliever. My guess is that the teams with the high powered offense had more wins by the starters, simply because they had more games with big leads early that allowed the pitcher (and manager) to chill and eat innings…but that’s a guess. It certainly stimulates some chicken vs egg thought, which is fun and healthy. Thanks for the insight.
I get your desire to have more context.
I added another column to the chart and included it in the article, with a hat tip to you. The addition was the percentage of wins by starting pitchers, which ranged from 46% to 68%.
It’s just one season’s worth of results and no doubt it would be better with at least one more year. Still, it seems like you better plan on your pen providing at least 30% of your team’s wins.
Thanks, and agreed. Now it makes me wonder even more….would it be better for the 2024 Mets to start Edwin Diaz for two innings every game possible but always with 2 days rest in between? Could he dominate for 100 innings and add more value….and more bullpen wins?
The gold standard for the Mets: 1986
78 wins for the starters .703 win%
3.16 ERA
30 wins for relievers .588 win%
2.98 ERA
I’d love that again this year. The starting staff: Gooden, Darling, Aguilera, Fernandez were all very young. Translate to Scott, Hamel, Vasil and Lucchesi? LOLLL
And check out John Gibbons’ 1.300 OPS!
Talking about team wins means finding the stats that most often lead to wins. SABR used Elias Sports numbers and figured out that the team that scores first wins the game about 67% of the time. That means being aggressive in the early innings with runners on base, ignoring moneyball’s bias against steals and sac bunts/flies. It also prioritizes power bats to get leads with early-inning HRs.
I’m curious to see how often the 2023 Braves and Dodgers scored first.
No stats on it, but I imagine an early lead pressures the losing team into mistakes or pressing which can improve a SP or reliever’s ERA.
The Moneyball bias against steals was misstated and has since been re-stated. In the generic sense, steal attempts were a losing effort over the long haul. But that was in an era in which everyone was running, and you had numerous players with merely average to above average speed stealing 15 – 30 bases a year while getting caught an equal number of times. 50% is a losing proposition because it not only creates an out, but it also removes an existing runner from the bases.
Since then, analytics has concluded that a 70+ % success rate is beneficial.