Does this pinpoint a specific card...

[moodinator123]

I saw Wade Boggs become available 18 games into a league of mine ( he has since been picked-up, so this is for general knowledge) and I looked through a few of his boxscores to see why someone would drop him so early ( he was hitting only .260-ish, but that's still a bit quick to drop someone). I found a boxscore that was pretty strange and wanted the opinions of a few of you vets...

Here is what I found. Boggs ( a lefty) was against Worrell ( a righty) in the 9 th inning. "HAL" pinch-hit for him with light-hitting Bob Melvin ( a righty) !!!!!!!!

Would this prove Boggs to be in his only "bad" year???

[YountFan]

It proves nothing except the donnot pich hit vs lefties was not checked. Boogs can lean way right and if HAL thought he needed more power, or what ever, he will pinch hit. Now it may smell like his bad year which is why it may have been dumped, but proof...no.

[moodinator123]

But Boggs was hitting against a Righty pitcher (Worrell), so since he is always better versus righties...and since Melvin has even less power and is a righty too...why on earth would HAL pick that match-up. It boggles the mind

[Paul5757]

Boggs has a lot of negative clutch in his second-worst year, negative clutch in some other years, and Worrell (as I recall) has an extreme reverse-righty year. Melvin, as I recall, also has a reverse-righty year. Can't recall if he has positive clutch.

[Outta Leftfield]

A lot of what HAL does doesn't appear to make sense! :D

It would seem to take Bogg's worst year in spades to make him worse than Melvin vs an RHP.

The issue of whether HAL is using knowledge of the card has been much debated--but I think we can now get to the bottom of the question. It seems to me that if we checked on some players who we've dropped in prior seasons at least in part because of this pinch-hit thing and who were later picked up by another team and held till the end of the year, we'd have an idea if the pinch-hitting is based on knowledge of the card--or not!

If we could get a critical mass of these cards, maybe we could deduce an answer to the question.

One player I've got my eye on is Ernie Whitt in a current league. He's a lefty hitting catcher with a pronounced platoon advantage vs RHP in most years. I dropped him in part because HAL was pinch hitting for him against RHP with righty hitters. [b:58cbdb053d]If HAL KNOWS[/b:58cbdb053d] this would be a pretty good sign of it. He had a .258 OPS for me in just 11 AB (I didn't give him much chance!) He was picked up and is still on another team where he's posted a .641 OPS in 214 AB--right in line with his worst season.) We're past the drop deadline so we'll know the year on this player after the playoffs. All of his other years are strong against RHP. So I would tend to think that if it's his worst year, [b:58cbdb053d]HAL knows![/b:58cbdb053d] And if it's one of the other years, then [b:58cbdb053d]HAL has no frickin' clue[/b:58cbdb053d] what he's doing.

I'll report back when the playoffs end. :wink:

[Paul5757]

When I first started in the 80s, I e-mailed TSN and asked if HAL knew the card. Their response was yes. Think about it too: If substitutions were random, you'd expect more of them.

The real question is why HAL makes subsitutions: Does he look at OBP only, OPS only, and does usage come into play at all. (Defense, as we all pretty much know, doesn't play a part!)

And you can't look at just "HAL PH for Boggs vs. a RHP." You have to look at who the RHP was, then do the math behind the situation. Here, I'd like to know what year Worrell was, and whether it was a clutch situation. You could figure out the hit, walk chances if you knew what year Worrell was.

HAL does some off-the-wall things, but I would bet for most situations, there is at least some "logic" for it. We just don't know what the "logic" is.

[Outta Leftfield]

Thanks for the info, Toi San, on TSN's reply about HAL. And you're asking all the right questions.

I'm beginning to think that with relief pitching decisions, HAL looks at ERA and ERA only--in terms of the specific seasons of the RPs. No reference to platoon advantage. Then there's a tiredness element to be considered. Then there's a "let's spread the work around," element. Then there's an element of "fuzzy logic," designed to prevent the computer from doing the same thing every time. But HAL doesn't know if it's the last game of the finals, so fuzzy logic may kick in in a randomized way just when you least want it too.

This whole set of factors produces a not completely predictable pattern, but it means the real go-to guy will be the guy with the lowest ERA, regardless of other circumstances. I'm suspecting this because at the beginning of one of my current seasons, HAL treated Tom Burgmeier as my go-to guy very persistently, until I put on a whole lot of controls. I think his card is one with a very low ERA (2.00 or 2.29), but someting like a 7L or 9L platoon. Lefties are hitting .117 against him while righties are hitting .298.

I had other relievers who were really much better, overall, despite having ERA's just a few points over Burgmeier, but until I set the controls really hard to keep him away from righties, Burgmeier got into almost all the key situations. I haven't dumped any of my relievers on this team, so when the season is over, I may be able to draw some firmer conclusions.

[Paul5757]

The fuzzy logic part makes sense, as it would explain some of the hiccups HAL has from time to time.

Have you played the CD-Rom game? The reason I ask is because there's a nifty little pie chart that can be used to show the out vs. safe % in a particular matchup. That leads me to believe HAL is most concerned about OBP for a particular matchup. HAL will calculate the safe chances from a given matchup according to the cards (e.g., Boggs vs. Worrell, not Boggs vs. any RHP)---factoring in balance, fatigue, and clutch---then he decides whether to make a change for that matchup.

The bigger question is what triggers a HAL move, not who HAL uses. In your case, I'd be willing to bet that when that trigger was pulled for HAL, Burgmeier was the best chance at getting ONE out. Unfortunately for you, that trigger isn't getting pulled for the second hitter, for whatever reason.

[YountFan]

I agree, I forgot Worell is a RHP, that it points to a bad Boggs year, but it doen't not prove it.

[Outta Leftfield]

[quote:d64179d2b8="Toi_San"]The fuzzy logic part makes sense, as it would explain some of the hiccups HAL has from time to time.][/quote:d64179d2b8]

Yes, when I thought of fuzzy logic, it seemed to explain a lot. For example, I've seen a couple of teams where the owner was apparently not at the controls and HAL was setting the daily lineups. Every lineup was screwy--yes--but they varied from day to day. That would suggest fuzzy logic, as would various other inconsistencies.

[quote:d64179d2b8] Have you played the CD-Rom game? The reason I ask is because there's a nifty little pie chart that can be used to show the out vs. safe % in a particular matchup. That leads me to believe HAL is most concerned about OBP for a particular matchup. HAL will calculate the safe chances from a given matchup according to the cards (e.g., Boggs vs. Worrell, not Boggs vs. any RHP)---factoring in balance, fatigue, and clutch---then he decides whether to make a change for that matchup. [/quote:d64179d2b8]

I've never played the CD-Rom game, but it's interesting about the pie chart and OBP. OBP makes sense as a HAL standard, because its a single number that covers most everything--in that way, resembling ERA. I guess the question is whether platoon is considered in the mix for pitchers.

[quote:d64179d2b8]The bigger question is what triggers a HAL move, not who HAL uses. In your case, I'd be willing to bet that when that trigger was pulled for HAL, Burgmeier was the best chance at getting ONE out. Unfortunately for you, that trigger isn't getting pulled for the second hitter, for whatever reason.[/quote:d64179d2b8]
You're right, the triggering mechanism in the key. In this case, I hadn't set any controls on HAL through the first 9 games. Burgmeier pitched in 4 of those games. In the first game he entered against Concepcion, a RH hitter, and pitched to 3 more righties (included an S), before being hammered out of the game. In the next game he pitched only to Garvey (R). In the next game he pitched to just Cooper (L) and Bonilla (S). In the final game he pitched to Reynolds (S) and then a whole string of 7 righies who flat out killed him. By the end of these 4 outings he had an ERA of 13.50 and had faced exactly 1 lefty and 13 righties (including 3 S's). It was then I decided I'd better set some controls. He's now on a very short leash and has a 3.35 ERA and 1.05 WHIP in 45.2 IP.

Anyway, why I suspect that HAL is looking at ERA only is that two of these outings began against RH hitters. Another began against an S, who obviously batted right. Only one began against a lefty--the only lefty (Cooper) that Burgmeier faced among 14 batters. I have four other RHP relievers who all murder righties. I strongly suspect that this is either his 2.00/9L or 2.29/7L season. At the end of the year we'll know a bit more, though maybe nothing for certain. I admit it would be odd for HAL to consult platoon advantage with pinch-hitters but not with RPs, but from what I'm seeing here it appears that it might be the case.