OhMrScottyTrav06 wrote:Experience may factor into the equation as well... as the Indians and Blue Jays (both were terrible in one-run games) were very young. The Red Sox, Yanks, Angels, White Sox are all experienced teams and thus, had the good records in one run games...
Correlation is not causality, though. Be careful of assuming that Eventuality A and Eventuality B are causally connected simply because they occur together often.
Your wisemen don't know how it feels to be thick as a brick...
Going to split a few hairs quietstorm, but I'm pretty much in agreement with what you wrote.
quietstorm wrote:"Wow. To continue with baseball: Every batted ball has a set of probabilities governing it. Aside from very minimal control of direction due to swing speed/timing, the batter has no ability to control where the ball goes. As such, and considering the defense, it may or may not fall in for a hit. Because of this, every batted ball of Velocity A has roughly the same chance of falling in for a hit, given that the defensive team remains the same (no stamina loss, etc.). What makes a good hitter is hitting more balls, yes.
Close but not quite. What makes a good hitter is not simply hitting more balls, its hitting more balls at Velocity X which have a higher probability of falling for a hit than a ball hit at Velocity A. There is much more than just a very minimal amount of control.
However, it's still always a probability issue. If you're a .333 hitter, every at-bat gives you a one in three chance that you'll end up on base (or hitting a home run). You can't control that any more than I can control a coin flip -- it's going to come out relatively even, given enough chances, but that doesn't mean that it's not going to land heads fifty times then tails fifty times. Every time you swing, it's the same probability.
Again close, but not quite. The game is still a batter - pitcher matchup so every AB for a .333 hitter is not a 1/3 chance for a hit. The batter pitcher matchup is much better defined by the log5 formula. Like I said earlier, its splitting hairs since over a very large set of data the ratio of "ace" pitcher matchups and "scrub" pitcher matchups tends to even out. After spending the last nine months pounding away at simulator code I just couldn't stop myself from pointing it out.
I think I said I was assuming static conditions, but I'm not certain. I was, if I didn't say so -- so the issue of ace/scrub is moot. Assume league-average static situation.
As for the velocity issue, I'm aware of that, but the more variables I put into something like this, the less useful it becomes for a broader, cross-sport purpose. It actually came from a poll I posted asking if a .500 team with a +2 scoring differential on average was better than a 1.000 team with a +1 differential in every game.
But, you're right, velocity is important to the equation in baseball, though good players have higher contact rates as a general rule.
As as for the amount of control a player has on the ball... as I said, the control is really specific to swing speed/timing, and potentially angle. But any control of the sort, "Hit the ball on the top left to make it go down and right" is out of the question, because of the speed of the pitch.
Your wisemen don't know how it feels to be thick as a brick...
I'm not sure if anyone has summarized the points, but the key things here are:
1. A team's record in one-run games does, in part, reflect their ability.
2. However, a team's records in one-run games are also influenced substantially by the lucky breaks of the game.
3. As a result, teams that generally perform very well or very poorly in one year, tend not to repeat that performance (in some sense, it's just normal regression to the mean.
4. Furthermore, a better predictor of a team's overall record next year, is based on looking at their performance measured by run differential and quality of opponents, rather than their actual record for the prior year.
5. So, looking at the 3rd order Pythagenports tends to be a better predictor.
I think one of the problems with the one-run differential approach is that it fails to take into account the quality of a bullpen. this is why bullpens are so important in baseball, so that a team can hold onto a one-run lead late in a game. Look at the teams with perhaps the best bullpens in the AL: Anahiem and Chicago. Those two teams, not surprisingly, made it to the ALCS. Sure, some of it is luck, but some of it is also good arms late in a game stepping it up to silence bats that can feast on lesser-quality starting pitching.
RedSoxdominate wrote:I think one of the problems with the one-run differential approach is that it fails to take into account the quality of a bullpen. this is why bullpens are so important in baseball, so that a team can hold onto a one-run lead late in a game. Look at the teams with perhaps the best bullpens in the AL: Anahiem and Chicago. Those two teams, not surprisingly, made it to the ALCS. Sure, some of it is luck, but some of it is also good arms late in a game stepping it up to silence bats that can feast on lesser-quality starting pitching.
quietstorm wrote:It actually came from a poll I posted asking if a .500 team with a +2 scoring differential on average was better than a 1.000 team with a +1 differential in every game.
The Green Bay Packers are currently asking themselves the very same question...
RedSoxdominate wrote:I think one of the problems with the one-run differential approach is that it fails to take into account the quality of a bullpen. this is why bullpens are so important in baseball, so that a team can hold onto a one-run lead late in a game. Look at the teams with perhaps the best bullpens in the AL: Anahiem and Chicago. Those two teams, not surprisingly, made it to the ALCS. Sure, some of it is luck, but some of it is also good arms late in a game stepping it up to silence bats that can feast on lesser-quality starting pitching.
But you're failing to take into account the possiblity that teams had bigger leads before winning a game by one run. What if most of their one-run wins happened after their bullpens gave up three or more runs in the last two innings? Then it's not an issue of bullpen quality.
Likewise, what if most of the one-run wins are a result of the other team's bullpen absolutely collapsing at the end of the game? (I realize this means one team is likely better than the other, but you have to look outside the individual matchup, to, say, a game between two other teams, one evenly matched with the winner of the first game, but the other with a better bullpen than the loser of the first game. Different opponent strengths and weaknesses are going to significantly affect a team's record, and the amount of games they play against each other is, by-and-large, a product of luck.)
Mookie4ever wrote:Over a 162 game schedule there is going to be no such thing as luck. Clutch is a more valuable concept than luck.
Except that clutch isn't quantifiable and thus we cannot assume that it exists. I'm sure that's what you meant to say, right?
Your wisemen don't know how it feels to be thick as a brick...
Mookie4ever wrote:Over a 162 game schedule there is going to be no such thing as luck. Clutch is a more valuable concept than luck.
Nonsense.
"Know what the difference between hitting .250 and .300 is? It's 25 hits. 25 hits in 500 at bats is 50 points, okay? There's 6 months in a season, that's about 25 weeks. That means if you get just one extra flare a week - just one - a gorp... you get a groundball, you get a groundball with eyes... you get a dying quail, just one
more dying quail a week... and you're in Yankee Stadium. "
- Crash Davis, In Bull Durham
Statistically, it takes 3-5 seasons of data to get most baseball results down to the point where "luck" plays an unimportant role.
Cafe Home
Fantasy Football
Fantasy Basketball
Fantasy Hockey
by quietstorm » Sun Dec 11, 2005 2:23 am 
