Jump to content

Providing independent coverage of the Minnesota Twins.
Subscribe to Twins Daily Email

The Forums

Trading for Price and Betts

Minnesota Twins Talk Today, 04:57 AM
Good Morning,   Per the MLBTR article linked, how would y'all feel about trying to swing a deal with the Red Sox for Mookie Betts an...
Full topic ›

2020 Baseball America Top 100 Prospects

Twins Minor League Talk Today, 02:52 AM
The Twins got 6 players in the 2020 Baseball America Top 100 prospects rankings.    https://www.baseball...100-prospects/...
Full topic ›

Astros Being Investigated for Using Tech to Steal Pitchin...

Other Baseball Yesterday, 06:01 PM
I can no longer say I want the Twins to emulate the Astros. Bush league maneuver to use cameras in order to steal signs.
Full topic ›

Fangraphs ZiPS Projections:

Minnesota Twins Talk Yesterday, 05:50 PM
https://blogs.fangra...innesota-twins/   A lot of interesting stuff for those who like projection systems. One quote that stood out...
Full topic ›

World Series or Bust

Minnesota Twins Talk Today, 02:56 AM
Looks that way, doesn't it?   We are clearly built for a bright overall future, but Cruz and Donaldson are at the tail end of their...
Full topic ›

The Stew Review: Imaginary Innings

As an official scorer, I enjoy reconstructing an inning after an error or passed ball to see if any runs that score are earned or unearned. Next to my regular scoresheet, with the play-by-play as it happens, I start an alternative-fact account on a separate notepad.

Errors fall into two primary categories: missed-out errors, which also create or extend the life of a baserunner, and advancement errors, which do not affect the number of outs or the number of baserunners, only the placement of the baserunners.
Image courtesy of © Jordan Johnson-USA TODAY Sports
  Stew Thornley is one of the Twins official scorers and a baseball historian. He will occasionally provide insight to the decisions official scorers make. If you have any questions you would like him to address in a future story, you can ask them in this Twins Daily thread. You can also read more from Stew at StewThornley.net.  

Missed-out errors are usually straightforward in reconstruction. If a player reaches base on an error, such as a shortstop booting his ground ball, and scores, his run is unearned. In addition, any runs that score after two out are unearned, at least to the team and also to the pitcher if he was in the game at the time the error occurred.

Advancement errors, such as an outfielder fumbling the ball after a hit, are more likely to require a judgment call if a run scores. Also, a run that scores because of an advancement error may be unearned but may become earned depending on subsequent events.

When an error or passed ball occurs, scorers should consider the impact after each ensuing play.

Consider this sequence:
  • A runner on first steals second and continues to third on an overthrow by the catcher. Situation: runner on third who would be on second without the error.
  • The runner from third scores on a sacrifice fly. The run is unearned at this time although it could become earned depending on subsequent events. The scorer should determine at this time if the runner would have advanced from second to third on the fly. Even though the runner scored from third, it does not mean he could have gone from second to third on the fly. If the fly was of medium depth to left field, the scorer may determine that the runner would still be on second if not for the error. If the fly was to deep right field, the scorer may determine that the runner would have advanced to third on the fly. (If the scorer is unsure, the benefit of the doubt should be given to the pitcher and the judgment made that the runner would still be on second.)
  • The next batter singles. If the scorer’s judgment was that the runner would have advanced to third on the fly ball in the preceding play, then the run that scored on the sacrifice fly becomes earned. If the judgment was that the runner would have remained at second on the fly, then the scorer must make another judgment on whether he could have scored from second on the single. This judgment should be made at the time of the single and tracked as part of the inning being reconstructed if the error had not occurred.
  • The remaining batter or batters complete the inning by making outs with no plays occurring that would have created additional advancement by the runner.
In determining if a runner would have been able to score from second on a single to the outfield, as noted above, the scorer may consider many factors. Some obvious ones are the speed of the runner, the strength of the outfielder’s arm, and the nature of the hit. Situational factors should also be considered:
  • Number of outs when the single occurred—the difference between two out and fewer than two out. With two out a runner can get a better jump on a batted ball. He does not have to hold up to see if a line drive or fly ball is caught or hesitate to see if a ground ball makes it through the left side of the infield. With fewer than two out, the runner sometimes has to hold up, and it is more likely that it can be considered that he would not have scored on the hit.
  • Likelihood of the third-base coach sending the runner home. This can be dependent on the number of outs as well as other factors, such as the score of the game. With two out, it is more likely that a team will take a chance and send the runner home. With one out, the coach may be more cautious and likely to hold the runner, determining that there is still a good chance he will be driven in by the batters who follow. With no outs, this is even more likely.
  • If a team is trailing by a large margin, it is less likely to take a chance. Therefore, the scorer has more reason to assume that the runner would not have been waved home. This can be the case even in a two-run game if it is the potential final inning for the batting team and the runner in question is not the tying run.
The inning has to be reconstructed based on what follows, even if logic dictates that the inning would have played out differently: a batter singles and goes to second when the outfielder lets the ball by him. The next batter is intentionally walked. The next batter homers. All three runs are earned. We know that the intentional walk would not have occurred if not for the error; however, intentional walks are treated the same as any other walk in reconstructing an inning.

Try this one: with one out, a batter reaches first on an error. The next batter triples, scoring the runner (unearned). The next batter hits a routine grounder to the shortstop, who throws home late trying to get the runner coming from third. Can the scorer assume that, without the error, there would have been two outs and the shortstop would have thrown to first for the final out?

No. The plays and actions by fielders that occur after an error must be assumed when reconstructing an inning.

However, a situation could arise where an out could be assumed if it does not assume any decisions made on the part of fielders. Example: With fewer than two out Simmons singles and advances to second when the center fielder bobbles the ball for an error. Torriente strikes out on a pitch that gets by the catcher. The wild pitch allows Simmons to go to third and Torriente to reach first. If not for the error, Simmons would have been on first, and Torriente—with first base occupied and fewer than two out—would be out automatically. An out on the strikeout could be assumed in this situation.

I’ve never gotten heat on the reconstruction of an inning. Give the pitcher the benefit of the doubt when making judgments (benefit of the doubt, not an outright gift). The team in the field is usually happy, and the team at bat doesn’t care. However, what happens if the pitcher affected is battling for the league lead in earned-run average with a pitcher on the other team? I’m hoping to avoid that scenario.

  • hybridbear, Broker and MN_ExPat like this

  • Share:
  • submit to reddit
Subscribe to Twins Daily Email

Subscribe to Twins Daily Email


Old Twins Cap
Aug 16 2017 10:02 PM

Thanks for this article.


It has always struck me as hypothetical when scorers reconstruct innings and assume that hits and plays would have been largely the same as before an error.


It many ways, inning reconstruction is akin to a cross between alternate reality and projection, based on a single sample size.


All this to preserve ERA as a comparative measure of pitching performance.


It really lacks in validity at the end of the day and inning.


I would bet a different measure emerges, more grounded in pitching metrics as to velocity/break/location that assess a pitcher's qualities and not the incidental results produced -- or not produced -- during an inning when expected defensive play has not been the norm.


For now, can we measure Actual ERA vs. Attributed ERA?

Thanks for the write up Stew...


Not to sure about that Cap. While it isn't a perfect system, to say it lacks validity is a wee bit harsh. Subjectivity is a large part of baseball, however, an experienced scorer who knows the players and their abilities and is skilled at reading the nuances of the game can usually make a pretty accurate read of the situation.  


Does that mean we are always going to agree with that decision, just like blue's decisions behind the plate? Not even remotely close. Like Stew said in the article, most decisions about scoring are fairly straightforward.  


Those tough or even bizarre instances though... the ones that make you turn your head to the side like your dog does when he is confused (happens a lot to mine) and you say to your self "WHAT.THE.*****!?!? WHAT THE.... WHAT WAS HE THINKING???" Those are what make baseball what it is.


I've score my son's games, and even that is a thankless job (If you've never actually scored a game, it's way tougher than it looks). I for one am thankful that folks like Stew here can do it and do it well.     

Aug 17 2017 07:12 AM
This is one of the many reasons to love baseball: inning reconstruction.

This is great stuff! I love your contributions!

    • IndyTwinsFan likes this

Maybe no one will wonder about this, but I feel like mentioning that giving the pitcher the benefit of the doubt is not simply one scorer's opinion, or even the collective wisdom of the guild of scorers down through the years, but is written explicitly in the rulebook.

Aug 17 2017 12:01 PM

Great article again, as always...


Stew said: "The inning has to be reconstructed based on what follows, even if logic dictates that the inning would have played out differently."


I assume that statement applies in the following situation, but I've always wondered about that as it applies to bunts. 

  • Buxton singles
  • Adrianza strikes out.
  • Castro grounds to 2B, but SS drops the ball on the force, resulting in 1st and 2nd with 1 out (should be runner on 1st with 2 outs)
  • Ervin Santana is up, so he bunts for the second out, moving the runners to 2nd and 3rd. (In the reconstruction, a bunt out would be the 3rd out, but he wouldn't be bunting in that situation. After all, it's Santana -- he would have slashed a double down the line!)
  • Dozier singles and both runs score.
  • Grossman strikes out.

I assume that would be two unearned runs, but the notion of reconstructing the inning with the bunt doesn't make much since there should have been two outs and Santana wouldn't have been bunting. 


What is the explanation on that one? Is it the claim that Santana was actually bunting for a hit? 

What a great article! Thanks for sharing your perspective!

Similar Articles

by Sarah , 09 Sep 2019

by John Bonnes , 21 Jan 2018

by stewthornley , 04 Jun 2017

by Sarah , 17 May 2016