How Bad Can You Be?

The definition of the word "luck" is open to a huge amount of personal interpretation.

I have four teams this season who won 100+ games. Those four teams had an average .601 winning pct in one run games (116-77). My four worst teams had a .466 winning pct in one run games (95-109). Because my managerial style is somewhat consistent over teams, I could attribute the variance to "luck". I would like all my teams to win 90 games, but unfortunately some win more than they need (>100) and some win less than they need (<80). I could use the word "lucky" to describe these outliers.

However, to be technical about it, it would probably be safest to say there is no such thing as luck. Things like three or four consecutive home runs could easily be interpreted as lucky too, however, there would be an assigned likelihood of these occurrences actually happening. Luck then might be a game ending flyout resulting in a game-winning hit because a Sharknado prevented the outfielder from making a routine catch.

In the NHL, almost half of all hockey games are decided by one goal. The parity of (most) teams is obvious, and this statistic speaks toward that. SOM Online is the same thing. Parity across the board is very strong. So for me, personally, I subjectively attribute luck with winning one run games. The point to be made is that I believe two teams are considered more equal the closer the final score is. For me and for the purposes of this conversation, luck doesn't happen in the unlikeliest of events (three consecutive home runs), but in the fewest possible events (walk-off home run).

I think this describes what Bill James and what most of the rest of us are articulating. If anyone else has a broader definition of luck, they're welcome to it. However, there is no official objective definition of what luck is, so this whole dialogue is subject to a lot of personal interpretation.

* * * * * * * *

"How much easier it is to be critical than to be correct." - Benjamin Disraeli

Take one-run games. People talk about the « character » it takes to win one-run games. Bill says that they involve a huge amount of luck – and that that may be the only safe statement that can be made about them.

This is interesting, but James doesn't explain why 1-run games involve "a huge amount of luck." So, it doesn't counter my argument. I have huge respect for the man, but even he has to support his arguments to me.

Also, remember I noted luck does not factor the same way in MLB games as it does in SOM games:

Also, luck's presence in MLB is significantly different than in Strat-o-matic. Luck is not the same in SOM as in real life. In MLB, luck can make its presence known through a myriad of natural phenomena: freak physical ailments, missed signs, bad calls etc. In online SOM, luck is isolated by the rolls of the dice and HAL's algorithms, so it can be better isolated and anlyzed. However, even in SOM, luck--i.e. unlikely occurrences/dice rolls--is as likely to be present in games with high-scoring margins as it is in one run games.


So, since James is talking solely about MLB luck, it doesn't apply to or counter my SOM "luck" argument.

The definition of the word "luck" is open to a huge amount of personal interpretation.

The definition of "luck" is open to interpretation. However, the definition I used is consensually accepted. I stated that "luck" is usually considered present in events countering probability, and it is. Most people don't apply luck to probable events.

I do, however, think it would be beneficial if we discussed our respective definitions of "luck." I'm certainly not opposed to amending mine.

I have four teams this season who won 100+ games. Those four teams had an average .601 winning pct in one run games (116-77). My four worst teams had a .466 winning pct in one run games (95-109). Because my managerial style is somewhat consistent over teams, I could attribute the variance to "luck". I would like all my teams to win 90 games, but unfortunately some win more than they need (>100) and some win less than they need (<80). I could use the word "lucky" to describe these outliers.

With his highly broad applications here, J-Pav could attribute any variances here to luck. That doesn't make them evidence of luck. He also gives no definition to the word "luck." The word may be up to personal interpretation, but if he's going to use it effectively in any way, he needs to give his actual definition of it.

However, to be technical about it, it would probably be safest to say there is no such thing as luck. Things like three or four consecutive home runs could easily be interpreted as lucky too, however, there would be an assigned likelihood of these occurrences actually happening. Luck then might be a game ending flyout resulting in a game-winning hit because a Sharknado prevented the outfielder from making a routine catch.

In the metaphysical sense, J-Pav may be correct. All natural phenomena are natural phenomena. However, our language and culture does recognize the dynamic of luck, so functionally, he's not. Luck is officially defined as: "the things that happen to a person because of chance." That is the culturally accepted definition, and the denial of it does not negate that acceptance.

In the NHL, almost half of all hockey games are decided by one goal. The parity of (most) teams is obvious, and this statistic speaks toward that. SOM Online is the same thing. Parity across the board is very strong. So for me, personally, I subjectively attribute luck with winning one run games. The point to be made is that I believe two teams are considered more equal the closer the final score is. For me and for the purposes of this conversation, luck doesn't happen in the unlikeliest of events (three consecutive home runs), but in the fewest possible events (walk-off home run).

First of all, J-Pav says there is no such thing as luck, then says luck occurs mostly in one-run games. That's entirely contradictory. There's either luck or there isn't. Secondly, as I said before, luck in no way is necessarily more prevalent in one-run games. If no lucky occurrences happened in the one-run game, the game wasn't decided by luck. A one-run game doesn't attract luck any more than any other.

I think this describes what Bill James and what most of the rest of us are articulating. If anyone else has a broader definition of luck, they're welcome to it. However, there is no official objective definition of what luck is, so this whole dialogue is subject to a lot of personal interpretation.

What J-Pav articulated was not what James articulated. Their reasonings were decidedly different. I do, however, agree that we all need to better determine what we all mean by luck. We don't have to all agree with each other on it, but it would be nice if we could come to some consensus.

"How much easier it is to be critical than to be correct."[/i] - Benjamin Disraeli

Firstly, nobody is ever going to be truly "correct." Post-structuralist philosophers have well shown that in the Humanities; and Heisenberg, Goedel, and String Theorists have shown that in the sciences. However, we should never stop debating, critiquing, or even sometimes criticizing; they help us all learn. J-Pav does all three, so the quote was appropriate.

As far as 1 run games goes--I think luck does play a larger part than most think---especially in on-line play---I have had 3 or 4 teams that were 1st or 2nd in the league run differential wise and not make the playoffs due to 1 run losses, one of the teams was 40 runs better than any other team in the league. Those are very vivid in my mind, especially one tournament team. I scratch my head sometimes, but in the end if I have a large +run differential--what can I do??

I've had a few poor/marginal teams make the playoffs that shouldn't have also, mostly due to being in a weak division--

At the moment exactly 1/3 of my teams made the playoffs, but only one champ of 9 teams in and 4 finalists. Since 4 of 12 make the playoffs I guess I'm par or dead average---those 27 teams averaged 85.4 wins---but I would rather have 11 teams in the playoffs and average 80 wins a team, or just 3 teams in but all champs. Success depends on perspective

As far as 1 run games goes--I think luck does play a larger part than most think---especially in on-line play---I have had 3 or 4 teams that were 1st or 2nd in the league run differential wise and not make the playoffs due to 1 run losses, one of the teams was 40 runs better than any other team in the league. Those are very vivid in my mind, especially one tournament team. I scratch my head sometimes, but in the end if I have a large +run differential--what can I do??

That is frustrating. It's happened to me many times, too. However, just because you suffered many one-run losses, it doesn't mean you necessarily lost those games because of luck. If you want to find out for sure, you need to go over those one-run losses and see how many of them were lost because of "lucky" rolls, not probable ones. Until you do, your belief those one-run games were lost because of luck is just speculation.

http://www.baseballprospectus.com/artic ... leid=18151

1 run games are largely determined by luck. not exclusively but largely.

http://www.billjamesonline.com/stats121/

Winning 1 run games, if primarily a skill, would show up in the next season's record too since many teams return similar rosters. But it doesn't. At all.

http://www.fangraphs.com/blogs/how-much ... run-games/

It's not that there is NO skill level in winning 1 run games over the course of a season. It's just that there's a lot more luck involved than with teams that have the ability to blow teams out. A single sequencing shift can alter the entire game, and that's not a predictable skill. Sequencing luck occurs in dice based games too so the real life vs simulation distinction that you're trying to make in this thread has no bearing.

I.E., single single single homer run out out out. 4 runs
vs

Home run single single single out out out 2-3 runs, with same component statistics. Simple sequencing difference, unpredictable from game to game, season to season.

http://207.56.97.150/articles/james_onerun.htm

And finally the great Bill James does explain how 1 run games "involve a huge amount of luck".

http://www.baseballprospectus.com/article.php?articleid=18151

1 run games are largely determined by luck. not exclusively but largely.

This experimental article not only doesn't prove one-run games are largely determined by luck. It doesn't even make that argument. It concludes:

"The Conclusion: To say that there is no skill in a team winning one-run games would be wrong. Teams that are good at scoring runs and preventing the other team from doing so will have a better chance at winning them."

This is hardly saying 1-run games are largely determined by luck.

Secondly, none of this applies to SOM games. MLB games aren't determined by dice rolls and logorithms; SOM is.


P.s. Even the author admits his theory rests on questionable analysis, so one should read it skeptically:
" There are plenty of problems with this analysis strategy, but the idea is that even over the course of a few years, a team maintains a lot of the same players (usually). It's not perfect, but it's about the best we have available"