I think
Barry Bonds gets intentionally walked too much. I
don't mean this as a fan of Bonds, which I am not.
Nor do I mean it as a fan of the Giants, which I also am
not. I don't even mean it as a general baseball fan,
which I actually am. No, I simply mean it as someone
interested in strategy and tactics. I think putting
Barry Bonds on base nearly once a game is a tactical error,
and I also think this can be proven mathematically.
This is what I hope to do.
It
seems obvious to me that the intentional base on balls
(IBB) is somehow not being used correctly. Some
numbers: Bonds has more IBBs (79) than any other
team. The next closest player in IBBs is Jim Thome,
and he has only 19. Is Bonds four times as dangerous
as Thome? I certainly don't think so. The
only active player with a higher slugging percentage
than Bonds--Colorado's Todd Helton--has only 10 IBBs
this year. Is Bonds eight times as dangerous as
Helton? Absolutely not. Clearly something
is wrong here. Either Barry Bonds is being walked
too much or a bunch of other players are being walked
too little. You already know what I think--let's
see what we can find out.
First, I decided to try something simple, to see if
I was on the right track. I compared the average
number of bases Bonds gets when he's allowed to hit
with the average number of bases he gets when he's intentionally
walked. For those familiar with statistics or
with professional gambling, those would also be called
his "expected" bases. Here are his stats through
Monday, July 26, along with some arithmetic:
40
Singles
15 Doubles
0 Triples
25 Home Runs
66 Non-intentional Walks
141 Outs
(40
x 1) + (15 x 2) + (25 x 4) + (66 x 1)/(40 + 15 + 25
+ 66 + 141) = 236/287
The
arithmetic tells us that for each time Bonds has a chance
to hit, his "expectation" is .822 of one base.
Of course, each time he gets an IBB, he gets one full
base. From this information, it looks like Bonds
provides less value when he actually has to swing the
bat. So far, so good. Still, this method
seemed a little simplistic to me, so I decided to try
something more sophisticated.
My
next step was to compare IBBs with non-IBBs using something
called Extrapolated Runs (XR). XR can be calculated
using an equation developed by Jim Furtado (the equation
can be viewed here--don't
ask me how he derived it, it involves too much math,
even for someone like me). Basically, it's a way
to give an expected run value to everything a batter
does, regardless of whether a particular event produces
any actual runs. For instance, in the XR equation,
a home run has a value of 1.440 runs. Of course,
any random home run can score anywhere from one to four
runs, depending on the number of people on base.
But over the long run, the expectation is that a home
run will be worth 1.440 runs. Along these same
lines, everything else has expected values as well.
Here
are Bonds numbers again, along with the respective XR
values:
40
Singles--(.50)
15 Doubles--(.72)
0 Triples--(1.04)
25 Home Runs--(1.44)
66 Non-IBBs--(.34)
79 IBBs--(.25)
141 Outs--(-.09)
And
here's another bit of arithmetic:
(40
x .50) + (15 x .72) + (25 x 1.44) + (66 x .34) - (141
x .09)/(40 + 15 + 25 + 66 + 141) = 76.55/287
That
means his XR per non-IBB plate appearance is .267, while
his XR for an IBB is .25; in other words, according
to this info, Bonds provides slightly MORE value when
he has to swing the bat. Of course, if you looked
at Furtado's equation you saw there were a few other
things to consider--strikeouts, sacrifice flies and
grounding into double plays--but overall they had a
minimal effect on Bonds' final numbers, so I left them
out. You know, just to keep things simple.
Since
I ended up with conflicting results, I decided that
the best solution was to set up a distribution table
of all possible scenarios. Then I could see what
would actually happen if Bonds was walked vs. what would
happen if he was allowed to hit. Next time, I
will describe my methodology, as well as take you through
one specific scenario: 9th inning, two outs, nobody
on base. Should Bonds be walked? Tune in
to find out.
Martell
can be reached at martell@babblog.com.
|
