THE BOOK--Playing The Percentages In Baseball

Ummm, with Mays speed...hitting a triple would be ideal as a sac fly could bring him home. Hittin’ a double means a single could bring him home. Staying at first means a double play ball is more likely since there’s a force out available. That’s not a non-stat-watching-batter, that’s stupid baseball!

I don’t believe Mays was a stupid baseball player.

Can you check to see how many times Mays hit a single when batting in front of McCovey and compare it to how many times he hit doubles & triples in front of McCovey? I’m very curious about this.

PS. Can you also check to see how many double plays were hit into when Mays was on 1st?

Why don’t you check?

From 1959 to 1971 Mays hit 163 2B/3B in front of McCovey and 570 singles.  Now, how many PA’s he had where McCovey was in the on-deck circle is very difficult to determine, but my best estimate is somewhere around 3000…

So, 22% of his non-HR hits were extrabase hits with McCovey batting.

Without Mac, he was at 26.5%.  The difference is 4.5%.

One SD is .016, putting him at nearly 3SD from the mean.

@ MGL ... I don’t have a DB full of this stuff where I can just check it. I’d have to eyeball every boxscore in entire career. I know some people have this stuff in DB’s, so I figured I’d ask if Tango could check.

@ Ryan & Tango… hmmm, interesting.

I think they probably get the data from B-R PI, but I am not sure.

Now, if a hit is not a sure double - i.e., you might get thrown out at second, then being able to IBB the next batter, if that is indeed correct, changes the break even point for attempting to stretch.  Similarly, even if an IBB is not in order, you would be more likely to try and stretch with a singles hitter up next, especially with 2 outs. On the other hand, if opposing managers issue the IBB to the next batter considerably too often, then you actually might want to encourage them to do so by taking the extra base a little MORE often than is warranted, given everything else.

IOW, on the whole, I would think that the correct play is probably to just do what you would normally do.  And it is certainly NEVER correct to not advance when you know you are going to make it 100% of the time, as in stopping at first on a possible triple, sure double. 

My guess is that if indeed Mays had fewer doubles and triples with Mac on deck than otherwise, and it wasn’t a fluke, that the difference was mostly or entirely on close plays.  IOW, when Mac was not up next, Mays was probably much more likely to try and stretch a single into a double, whereas with him on deck, he rarely tried to do that, when he knew that he could be thrown out at second.  I sincerely doubt that he ever stopped at first on a ball that might have been a triple.  That would look awfully strange, if nothing else, and surely Mays would know that a triple is too valuable to concede even if they were to walk Mac.

Tango:  how are the McCovey and non-McCovey PAs distributed by year?  From 1959-64, Mays’ XBH% was 27.3%, but from 1965-71 it drops to 22%.  So some of this could be an aging issue.

Since Ryan has the data handy, he can report that.  If he can’t, I’ll run a query later…

And Mac’s big years were 63-70…

One must also take into account how often 1st base was occupied when Mays hit his singles and whether the runner ahead of him held up at second or not.

From 1959-1971, Mays had 1637 base hits, 406 for extra bases, or 24.8%, just about league average. As noted above, that did drop in the late 60’s.

From #4, he was 163 of 733 (22.2%) with McCovey next, so was 243 of 904 (26.9%) without McCovey next. At 26.9%, he would be expected to have 197 XBH on 733 BH, 34 more, or about 3 a season.

I am doing queries against a RetroSheet DB.

Here’s how his 2B+3B break down by year, when the next batter is McCovey.

1959 10
1960 12
1961 10
1962 6
1963 6
1964 12
1965 7
1966 23
1967 11
1968 19
1969 15
1970 15
1971 17
TOTAL 163

I am not really great at analysis, so I prefer to just give data ...  I would consider myself a “research assistant.” Or lab monkey...haha.

Anyway.  Here is how many 1B he had in front of McCovey, each year:

1959 22
1960 38
1961 30
1962 25
1963 33
1964 40
1965 40
1966 59
1967 36
1968 67
1969 65
1970 82
1971 33

If we divide the years in two (somewhat arbitrary), it looks like Mays did have a lower XBH ratio from 1959-64 when hitting in front of McCovey:  23%, vs. 27% overall.  From 1965 to 1971, with a sample about twice as large, there’s no difference (22% and 22%).  So if Mays did this, it appears it was only early in his career.  I’m skeptical.

Another check would be looking at May’s SBAs with and without McCovey batting.  If Mays was giving up 2Bs on purpose, he presumably wouldn’t then try to steal 2B and take the bat out of McCovey’s hands.

By batter (1959-1971)

Batter 2B+3B 1B
mccow101 163 570
cepeo101 109 261
hartj102 41 143
alouf101 18 46
kirkw101 14 28

--Bunch of people under 10 2B+3B--

This follows with #7-Is it possible to tell how many times Mays was thrown out at second stretching a single into a double? It seems unlikely that he would stop at first if he hit it to the wall, and had close to a 100% chance of making it to second, but if he was taking second on hits that were cut off in the gaps or down the line, where his chance of making it to second were only 70-80 percent, it would make his claim more reasonable.

"but if he was taking second on hits that were cut off in the gaps or down the line, where his chance of making it to second were only 70-80 percent, it would make his claim more reasonable.”

Right, as I said, the value of the extra base is going to be lower when you have a power hitter behind you and when they can walk that hitter (assuming that the walk is correct) to lessen the RE.  As well, the value of the out is going to be worse if it is the 3rd out.

One thing to look at is the number of outs.  If it is 2 outs, then to hold up when your chances of making it is less than 70 or 80% may be correct.  With less than 2 outs (and certainly 0 outs), it might never be correct unless your chance is less than 70% or so.

Bottom line is that there is a break even point, exactly the same as with a SB attempt, and that break even point depends on lots of factors, including the quality of the next batter, including the chance of him being IBB’d and whether that IBB reduces the RE or WE (as compared to not IBB’ing him).

The other bottom line is that if the chance of making second is 90% or higher (I am just throwing out a number), it can almost never be correct and if it is sure double/possible triple, then it is certainly not going to be correct.

Again, if it were correct, then it wouldn’t matter if he stopped or not, since the other team could just walk Mac anyway (with Mays on first).  So the first question I would have asked Mays was, “Well if you stopped at first to avoid them walking Mac, why didn’t they just walk him anyway?”

So, either Mays is mis-remembering or he was implementing a stupid strategy.

I also vaguely remember that he said that he would sometimes intentionally strike out early in a game in order to set up a pitcher late in the game.  Again, that cannot be correct.  I suppose it is possible with 2 strikes and 2 outs and no one on base or something like that.  But that seems like a tall tale to me.

80 year old people often have a knack for embellishing their accomplishments (not that Willie Mays needs to) and making them seem more interesting than they really were…