What do statistics really show?
| 7
|
by user Alex Holowczak
My Maths teacher said at the start of my second piece of Maths coursework this year that "Statistics don't prove anything." I think this goes along with that theory.
After reading a lot of Youkilis v Reyes articles, and seeing statistics used to prove one anothers case as winning, I'm not sure that statistics are the way to go, as I now hope to demonstrate (using baseball as an example).
Hits/Doubles/Triples/HRs (hence batting average)
Reflect Player's ability: Hitting the ball
Does not reflect Player's ability: You might have an division with poor pitchers, the NFC North of MLB Pitching. Lower quality pitchers = these stats increasing.
Walks (hence OBP)
Reflect Player's ability: Patience, leaving a ball missing the plate
Does not reflect Player's ability: Playing against pitchers with low accuracy increases OBP.
RBIs
Reflect Player's ability: Comes through when needed
Does not reflect a Player's ability: He may have a poor hitting team, and hence have no runners to hit in.
SBs
Reflect Player's ability: Good speed, awareness
Does not reflect a Player's ability: A pitcher in control and a catcher with a strong arm reduces the chance of stealing bases.
R
Reflect Player's ability: Good speed, awareness
Does not reflect a Player's ability: A bad team/good pitcher may provent anyone getting around.
It can even be turned around the other way...
Ks
Reflect Player's ability: Quality pitching
Does not reflect a Player's ability: Could be poor batting opposition.
ERA
Reflect Player's ability: Quality pitching
Does not reflect a Player's ability: Could be good/poor batting opposition.
All the statistics above are therefore flawed in some way, admittedly some more severely than others. But I think this shows that a players stats rely not only on the ability of the player, but the ability of the opposition.
Which seems to make sense.
Throwing this into the Youkilis-Reyes argument, I'm not sure stats can be used as persuasive weapons, as they play in separate divisions (separate Leagues for that matter), so can not really be compared in any meaningful way with statistics.
Taking this idea further, comparing players of different eras is even harder to do, as different opposition may make it easy to club 500HRs in a career in one era, but struggle to get to 200HRs in a different era.
This makes sense too, as pitching statistics are better in some eras than others.
So as for Youkilis v Reyes, I think everyone should agree to disagree, they are both great players.
N.B. Although if you could disect stats to see matchups against each pitcher, and compare them then, then you could get somewhere (hence, playing in the same division would help for comparison).
Date
Sat 06/24/06, 8:20 am EST
