Is this just a run of bad luck?

The Minnesota Twins have started the season with a piss-poor 2021 record. This has them in a deep hole in the AL Central, which they will need to dig out of to come close to offseason expectations. Despite their record being the actual result of the games, there is no logical reason they should be this bad, and their stats reflect this fact.

Pythagorean record is basically a way to measure what a team’s record should be, based on the numbers of runs they have scored and have surrendered. For the Twins, they entered play on Sunday with a Pythagorean record of 12-13. While that is certainly not great, its a far better outcome than their real record of 9-16. What does that mean though? Well, certainly there is an element of luck at play here, no matter what some of the stats that are considered a proxy for luck represents. More than anything, it means the Twins have been competitive, and have been losing winnable games. May I offer up the Alex Colome factor? While he certainly doesn’t belong in the ninth inning with a close lead right now, history suggests he should be much better than he has been. Even if this is his “new normal” there was no way the Twins could have predicted this outcome when they signed him.

BABIP is probably the best proxy available for “luck” as it measures how many balls in play actually turn into hits. A very low number means balls aren’t falling, while a high number means the team is missing gloves. The Twins hitters have a team BABIP of .278, which ranks 17th in the majors—while slightly below the median, its not indicative of a cellar-dweller. The actual rate stats don’t show a problem either—the Twins are top-ten in batting average and slugging percentage, and 16th in on base percentage. Nineteen teams have higher strikeout rates, although nineteen teams also walk more. Essentially, the Twins should be considered somewhere near a league-average offense, if not a bit better right now. So perhaps the pitchers are at fault?

BABIP against works sort of the same way for pitchers that BABIP works for hitters. The Twins, however, have an excellent defense, and the tenth-lowest BABIP against in the league at .271, so again, not a lot of bad luck going on there. The issue, however, isn’t balls that are in play, its balls that leave the park. The team ERA is 4.35, and the team’s FIP (which takes the defense out as a factor) is 4.45. xFIP, meanwhile, is a stat that neutralizes the home run rate, and shows the Twins pitchers should have an ERA of 4.09, if they could get the long balls under control. Further proving this point, the Twins’ 16.1 HR/FB ratio is worse than all-but-two teams, Cleveland and Atlanta, and equalled only by the Cubs. All three of those are 12-win squads, before Sunday’s action. So perhaps we found the culprit for the Twins record, and its not really a bad luck stat, but in a way, it might be.

The Twins have had the ignominy of giving up too many long-balls in their skid. Clearly the stats say so, but how much of that is luck? Well, here is a succinct summary from a site called Betting Pros, which means I don’t need to re-write it.

HR/FB is a stat that tracks the number of home runs a pitcher gives up relative to the number of fly balls they allow. Pitchers clearly do not want to surrender home runs, so a low HR/FB rate is ideal. So, where does the luck come in with the HR/FB statistic? While pitchers do have a level of control over the type of hit a batter will be able to put on a given pitch, controlling the speed and distance of a fly ball hit in play doesn’t exactly fall in the skill category. Some balls may be caught on the warning track. Others might carry to the seats for a home run. These variances are next to impossible for bettors to predict, but a poor HR/FB ratio would suggest that those balls have tended to leave the yard against a certain pitcher. The luck concept behind the HR/FB statistic is why general betting advice says to look to bet on pitchers that induce more grounders than fly balls because of the decreased amount of variance in play.

Sorting all Twins’ pitchers by HR/FB says a lot about how their season has gone. A quick glance to the left or right on this chart shows the high correlation between success and HR/FB rate. The sample is still small, but it makes perfect sense.

So overall, perhaps we can say that the Twins have had bad luck play a factor. They are four games below their expected record—if they were four games above, they would be 16-9 instead of 9-16, and tied with the Royals as the best team in baseball (again, before the game on Sunday.)

None of this luck conversation touches on the injury-or-illness issue either. The Twins have lost multiple players to those factors. Andrelton Simmons, Max Kepler, and several others have spent time on the COVID list, while Miguel Sano and Josh Donaldson have both spent time on the traditional Injured List. Byron Buxton and Nelson Cruz missed time with non-Covid illnesses as well. There is some element of luck in any injury, and even more so in who catches an illness, so we cannot discount that factor either—would this team have won a few more games at full strength? Its certainly possible.