Is HAL conservative tagging up from third?

A while back, I saw comments on here claiming that HAL was conservative on the basepaths. I believe there was specific reference to tagging up from third on a fly. I finally got around to investigating and am posting what I found for MLB and SOM. MLB data is from retrosheet play-by-play files. SOM data is from play by play data from leagues I've been in. In each of the SOM cases, the data is from 3 12-team leagues of the type indicated.

The data below shows that HAL may indeed be a little more conservative, but the conservatism may be justified because the runner is more likely to be thrown out in SOM compared to MLB. The ATG9-140 results look close to the older MLB data.

Flyout to OF
Run. on 3rd <2 outs
Outcome	MLB 1940s	MLB 1950s	MLB 1960s	MLB 1970s	MLB 1980s	MLB 1990s	MLB 2000s	MLB 2010s	MLB 2021	MLB 2022	SOM 2021	ATG9-80	ATG9-140
SF	74.4%	74.6%	74.5%	73.7%	76.4%	78.7%	78.1%	77.9%	76.4%	77.2%	70.2%	77.9%	74.9%
Hold	22.3%	22.0%	22.4%	22.9%	20.6%	18.6%	19.1%	19.7%	21.4%	20.3%	24.5%	18.1%	22.2%
Thrown out	3.3%	3.4%	3.2%	3.5%	3.0%	2.7%	2.8%	2.4%	2.2%	2.5%	5.3%	4.0%	2.9%

good stuff , I would think the 140 cap skews the result towards conservative results. Better outfield arms and managers are more likely to have less aggressive settings in their team strategy. I would be curious if the numbers hold up at 80m.

I would be curious if the numbers hold up at 80m.

80m is shown in the second to last column. I should probably pull more SOM data to see how stable it is.

Super interesting stuff. Honestly the thrown out seems low in both ATG and MLB (seems like perfect strategy would lead to more outs). But it's great that Strat is more or less mirroring real life.

barrmorris that is valuable data.
With a smallball team at low to mid caps I always feel like HAL is not aggressive enough. When I have 3 runs on 15 singles, then I ain't taking enough chances. I usually have to combat that by bumping up the stealing, which has a narrower success:risk envelope

barrmorris that is valuable data.
With a smallball team at low to mid caps I always feel like HAL is not aggressive enough. When I have 3 runs on 15 singles, then I ain't taking enough chances. I usually have to combat that by bumping up the stealing, which has a narrower success:risk envelope

Small Ball and stealing is why I'd love to have global team settings for each base. We could set very aggressive for 2b, normal for 3b, and Don't for steals of home.

Wow! That data is great! I haven't looked at this type much, but I can't believe the #s (that I never imagined). ~20% for "hold" percentage seems super low considering "Out %" is only at 2-4%. That's kind of hard to believe, but I did start by picking at whether all FBs to OF were valid---I'm unfamiliar with this type of data. 16-1 ratios for safe at home (with only ~5 parts being "stay") seems high, but this isn't PLAYS at the plate so morebelievable

Awesome stuff barrmorris!

Interesting comment by FrankieT about the 15 hit 3 run game

One thing to remember when interpreting this data is that not all of these flyballs are really baserunning decisions in SOM (and MLB, but harder to calculate). Fly A and Fly B are automatic SFs. Fly C is automatic hold. Only Fly B? results in a decision to be made.

Looking at card data, my crude estimates are that 62% of flyballs are A or B. 7% are C, and 31% are B?.

More crude calculations then get me to estimating that HAL tries for home on about 64% of Fly B?s and is out on 20% of those attempts.

All of this assumes that the "more baserunning decisions" max rule only applies to singles/doubles.


Edited to revise/correct estimates

A fast team should always be set to VERY AGGRESSIVE in this game. That will increase your attempts to score from 3rd on a flyball.

A fast team should always be set to VERY AGGRESSIVE in this game. That will increase your attempts to score from 3rd on a flyball.

Without any doubt. I wish there were a UBER AGGRESSIVE or LUDICROUS AGGRESSIVE setting