The Secret Formula 2007
by the splinter » Thu Oct 25, 2007 1:24 pm
Well what ever PBTR says is the secret for 07 I gotta believe him...he is the 07 stud without a doubt.
- the splinter
- Posts: 55
- Joined: Tue Jul 03, 2012 2:34 pm
by Mean Dean » Thu Oct 25, 2007 1:52 pm
I think it's an unsupported leap to go from "the teams that do best in one-run teams are not necessarily the best teams overall" to "overperforming your Pythag is mostly luck." The key IMO is bullpen -- and I'm not just saying that because it makes logical sense (which it does,) I'm saying that from observation of most of the teams that drastically over/underperform Pythag. Anyone can look them up themselves, and will notice right away a strong correlation to bullpen quality. So both things can be true: The best teams often do worse in one-run games because they don't necessarily have the best bullpens (bullpen being less important overall than SP and front-line hitting,) and yet, overperforming Pythag isn't "mostly luck," because a lot of it is the skill of the bullpen.
- Mean Dean
- Posts: 55
- Joined: Tue Jul 03, 2012 2:34 pm
by keyzick » Thu Oct 25, 2007 2:34 pm
I think platoon usage is another overlooked strategy, or maybe one that is taken for granted may be more accurate.
Up until the 07 season, I was strictly an autodraft, play-for-fun, wonder-why-my-teams-hover-below-.500, manager. A few things changed for me for some reason, starting with the 07 season:
1. Started paying attention to player/ballpark matchings ([b:252c4b4932]J-Pav [/b:252c4b4932]outlines its importance above)
2. Made sure to have 1's or 2's at 2b/ss ([b:252c4b4932]J-Pav [/b:252c4b4932]also writes about this, as well as cf...that's one position i've often used a 3 at though, so maybe another key area i can look at improving)
3. Most importanly, IMHO, started using platoons in anywhere from 1 to 3, or even 4 different positions, in order to maximize my matchups. This often results in using guys you wouldn't otherwise normally even consider, but finding the right mix for R/L matchups can be huge...often putting up combined stats comparable to some of the much pricier star players at the same positions.
J-Pav...have you ever looked more deeply into platoon usage? I wonder what kind of stats you might dig up there.
I've also deployed these same strategies in ATG3, along with ensuring I have a 1 at at least 6 of the 7 fielding positions (I don't worry about catcher), and have seen my results for '07 and ATG3 be far and above anything I did in previous sets. I learned (not as quickly as I'd like) in ATG3 that their are so many good players in the set, that a powerful offense can be easy to put together, so defense tends to be the difference maker.
On a side note to any newbies who might read this - getting involved in as many theme leagues as possible has also been a tremendous help. Seeing other managers' strategies, and playing against MUCH stiffer competition.
And finally, while I know many if not most managers do it already, I have to credit [b:252c4b4932]Geekor[/b:252c4b4932], [b:252c4b4932]Jeep[/b:252c4b4932], and [b:252c4b4932]visick [/b:252c4b4932], among other, for opening my eyes to really pay attention to total chances on player cards.
In hindsight, most of it seems pretty common sense and logical, but can easily be overlooked by the casual player. :D
Up until the 07 season, I was strictly an autodraft, play-for-fun, wonder-why-my-teams-hover-below-.500, manager. A few things changed for me for some reason, starting with the 07 season:
1. Started paying attention to player/ballpark matchings ([b:252c4b4932]J-Pav [/b:252c4b4932]outlines its importance above)
2. Made sure to have 1's or 2's at 2b/ss ([b:252c4b4932]J-Pav [/b:252c4b4932]also writes about this, as well as cf...that's one position i've often used a 3 at though, so maybe another key area i can look at improving)
3. Most importanly, IMHO, started using platoons in anywhere from 1 to 3, or even 4 different positions, in order to maximize my matchups. This often results in using guys you wouldn't otherwise normally even consider, but finding the right mix for R/L matchups can be huge...often putting up combined stats comparable to some of the much pricier star players at the same positions.
J-Pav...have you ever looked more deeply into platoon usage? I wonder what kind of stats you might dig up there.
I've also deployed these same strategies in ATG3, along with ensuring I have a 1 at at least 6 of the 7 fielding positions (I don't worry about catcher), and have seen my results for '07 and ATG3 be far and above anything I did in previous sets. I learned (not as quickly as I'd like) in ATG3 that their are so many good players in the set, that a powerful offense can be easy to put together, so defense tends to be the difference maker.
On a side note to any newbies who might read this - getting involved in as many theme leagues as possible has also been a tremendous help. Seeing other managers' strategies, and playing against MUCH stiffer competition.
And finally, while I know many if not most managers do it already, I have to credit [b:252c4b4932]Geekor[/b:252c4b4932], [b:252c4b4932]Jeep[/b:252c4b4932], and [b:252c4b4932]visick [/b:252c4b4932], among other, for opening my eyes to really pay attention to total chances on player cards.
In hindsight, most of it seems pretty common sense and logical, but can easily be overlooked by the casual player. :D
- keyzick
- Posts: 55
- Joined: Tue Jul 03, 2012 2:34 pm
by J-Pav » Thu Oct 25, 2007 3:25 pm
[b:ee2d742b79]Dean[/b:ee2d742b79]:
Are you quoting [i:ee2d742b79]me[/i:ee2d742b79] (did I say that somewhere?) or from somewhere else?? I think I have a response, but I'm not sure where you're coming from exactly.
[b:ee2d742b79]Keyzick[/b:ee2d742b79]:
Platoons are [i:ee2d742b79]extremely[/i:ee2d742b79] important to the way I set up; however, one or two at most is max for me. I think if you use too many, you open yourself up to knocking out a starter who is then replaced with the opposite hand reliever. The new guy is now looking at a lot of weak side hitters. I'd hate to knock out Halladay in the first, only to see Chris James come in and pitch seven innings like Cy Young (extreme example, but you get the point). HAL is extremely poor with situational batting and substituting players for match-up hitting.
But one good platoon, I give it a big [size=18:ee2d742b79]YES[/size:ee2d742b79]. Davanon/Nady, Edmonds/Logan and Matsui/Sheffield as well as several other combinations, have all given me 100 runs scored in tandem from the lower half of the order.
This is a huge advantage over the guys who don't use the whole roster to it's full capability.
And again, this speaks to salary construction and spending a lot of time thinking about those "inconsequential" guys on the bottom of your roster. Why use Hafner as a DH when you can use a sub $1.00 guy like Lopez, or a low budget platoon, and upgrade four other positions by two dollars each?
This goes back to choosing the right players for your park as well. I've never used him, but a guy like Branyan in The Cell seems like a pretty good idea to me.
Are you quoting [i:ee2d742b79]me[/i:ee2d742b79] (did I say that somewhere?) or from somewhere else?? I think I have a response, but I'm not sure where you're coming from exactly.
[b:ee2d742b79]Keyzick[/b:ee2d742b79]:
Platoons are [i:ee2d742b79]extremely[/i:ee2d742b79] important to the way I set up; however, one or two at most is max for me. I think if you use too many, you open yourself up to knocking out a starter who is then replaced with the opposite hand reliever. The new guy is now looking at a lot of weak side hitters. I'd hate to knock out Halladay in the first, only to see Chris James come in and pitch seven innings like Cy Young (extreme example, but you get the point). HAL is extremely poor with situational batting and substituting players for match-up hitting.
But one good platoon, I give it a big [size=18:ee2d742b79]YES[/size:ee2d742b79]. Davanon/Nady, Edmonds/Logan and Matsui/Sheffield as well as several other combinations, have all given me 100 runs scored in tandem from the lower half of the order.
This is a huge advantage over the guys who don't use the whole roster to it's full capability.
And again, this speaks to salary construction and spending a lot of time thinking about those "inconsequential" guys on the bottom of your roster. Why use Hafner as a DH when you can use a sub $1.00 guy like Lopez, or a low budget platoon, and upgrade four other positions by two dollars each?
This goes back to choosing the right players for your park as well. I've never used him, but a guy like Branyan in The Cell seems like a pretty good idea to me.
- J-Pav
- Posts: 55
- Joined: Tue Jul 03, 2012 2:34 pm
by teamnasty » Thu Oct 25, 2007 5:25 pm
JPav: Thanks for this fun, well-thought out article.
I'm not sure that one can conclude that run prevention is more important than run scoring for 2007 based on your 18 top teams averaging 4th in pitching vs 5th in offense. One spot of difference over 18 leagues is not much difference at all, and a sample of 18 leagues is large enough to be important but not so large as to be conclusive. If anything, your numbers suggest that balance is important to your league winners as opposed to imbalance; i.e., your winning teams tend to be slightly/moderately above average in both run prevention and run scoring, as opposed to high offense/avg pitching or high pitching/avg offense.
When it comes to run scoring your numbers do suggest that winning teams pull further away from the pack in onbase percentage than slugging, which is consistent both with my experience and Beane/James' entire moneyball philosophy. Good teams recognize the importance of walks, bad teams underrate walks. That still holds true 30 yrs after James started writing his Abstracts. And another Jamesian insight; building for park; remains correct.
I'm not convinced that outstanding defensive ranges up the middle, as compared to other positive factors, has some extra special significance to building a winner in 2007. It is precisely because up the middle D is the conventional wisdom "cliched" way to win, to borrow your term, that those players are overvalued in autodrafts.
Finally, I think your 30/50 pitching/offense budget suggestion seems about right, but note that your suggested split and emphasis on starting pitching may not hold as true in non-DH leagues. Sticks on the bench and quality relievers gain value there.
Thanks again for the shared studies and insights.
I'm not sure that one can conclude that run prevention is more important than run scoring for 2007 based on your 18 top teams averaging 4th in pitching vs 5th in offense. One spot of difference over 18 leagues is not much difference at all, and a sample of 18 leagues is large enough to be important but not so large as to be conclusive. If anything, your numbers suggest that balance is important to your league winners as opposed to imbalance; i.e., your winning teams tend to be slightly/moderately above average in both run prevention and run scoring, as opposed to high offense/avg pitching or high pitching/avg offense.
When it comes to run scoring your numbers do suggest that winning teams pull further away from the pack in onbase percentage than slugging, which is consistent both with my experience and Beane/James' entire moneyball philosophy. Good teams recognize the importance of walks, bad teams underrate walks. That still holds true 30 yrs after James started writing his Abstracts. And another Jamesian insight; building for park; remains correct.
I'm not convinced that outstanding defensive ranges up the middle, as compared to other positive factors, has some extra special significance to building a winner in 2007. It is precisely because up the middle D is the conventional wisdom "cliched" way to win, to borrow your term, that those players are overvalued in autodrafts.
Finally, I think your 30/50 pitching/offense budget suggestion seems about right, but note that your suggested split and emphasis on starting pitching may not hold as true in non-DH leagues. Sticks on the bench and quality relievers gain value there.
Thanks again for the shared studies and insights.
- teamnasty
- Posts: 55
- Joined: Tue Jul 03, 2012 2:34 pm
by J-Pav » Thu Oct 25, 2007 10:50 pm
[quote:59a268f1ff="DeanTSC"]I think it's an unsupported leap to go from "the teams that do best in one-run teams are not necessarily the best teams overall" to "overperforming your Pythag is mostly luck."[/quote:59a268f1ff]
I'm not sure I understand, because I don't think I said (or at least meant) what you are paraphrasing. If you want to pursue it some more, you have to explain a little bit more what you're saying.
I'm saying:
1. Most of the time, the pythag projection is right on.
2. When it's not, the only thing which stands out to me as noticeably different is the results of one-run games: [i:59a268f1ff]usually[/i:59a268f1ff], good teams win more of these, but it's not [i:59a268f1ff]always[/i:59a268f1ff] the case.
3. The results of one-run games are more luck than anything else.
These are the conclusions reached in the three articles I linked, as I understand it.
[quote:59a268f1ff="DeanTSC"]The key IMO is bullpen -- and I'm not just saying that because it makes logical sense (which it does,) I'm saying that from observation of most of the teams that drastically over/underperform Pythag. Anyone can look them up themselves, and will notice right away a strong correlation to bullpen quality. So both things can be true: The best teams often do worse in one-run games because they don't necessarily have the best bullpens (bullpen being less important overall than SP and front-line hitting,) and yet, overperforming Pythag isn't "mostly luck," because a lot of it is the skill of the bullpen.[/quote:59a268f1ff]
I agree and disagree.
It makes logical sense to me too; however, how do you judge bullpen "quality"? I looked at plenty of teams and saw no correlation with the bullpen. I often use Valverde/Beimel as my closer combo with great success. Is this better or worse "quality" than a Papelbon/Wagner combo? The salaries are different, but does that mean something [i:59a268f1ff]objective[/i:59a268f1ff] with regard to quality?
The best way to think of this would be, [i:59a268f1ff]What combination of five relief pitchers would give you ten more wins than projected?[/i:59a268f1ff]
The answer, I believe, is that you can't forecast bullpen "quality" ahead of time, so I can't agree with what you're saying.
I'm not sure I understand, because I don't think I said (or at least meant) what you are paraphrasing. If you want to pursue it some more, you have to explain a little bit more what you're saying.
I'm saying:
1. Most of the time, the pythag projection is right on.
2. When it's not, the only thing which stands out to me as noticeably different is the results of one-run games: [i:59a268f1ff]usually[/i:59a268f1ff], good teams win more of these, but it's not [i:59a268f1ff]always[/i:59a268f1ff] the case.
3. The results of one-run games are more luck than anything else.
These are the conclusions reached in the three articles I linked, as I understand it.
[quote:59a268f1ff="DeanTSC"]The key IMO is bullpen -- and I'm not just saying that because it makes logical sense (which it does,) I'm saying that from observation of most of the teams that drastically over/underperform Pythag. Anyone can look them up themselves, and will notice right away a strong correlation to bullpen quality. So both things can be true: The best teams often do worse in one-run games because they don't necessarily have the best bullpens (bullpen being less important overall than SP and front-line hitting,) and yet, overperforming Pythag isn't "mostly luck," because a lot of it is the skill of the bullpen.[/quote:59a268f1ff]
I agree and disagree.
It makes logical sense to me too; however, how do you judge bullpen "quality"? I looked at plenty of teams and saw no correlation with the bullpen. I often use Valverde/Beimel as my closer combo with great success. Is this better or worse "quality" than a Papelbon/Wagner combo? The salaries are different, but does that mean something [i:59a268f1ff]objective[/i:59a268f1ff] with regard to quality?
The best way to think of this would be, [i:59a268f1ff]What combination of five relief pitchers would give you ten more wins than projected?[/i:59a268f1ff]
The answer, I believe, is that you can't forecast bullpen "quality" ahead of time, so I can't agree with what you're saying.
- J-Pav
- Posts: 55
- Joined: Tue Jul 03, 2012 2:34 pm
by J-Pav » Thu Oct 25, 2007 11:31 pm
[quote:3a97672de3="teamnasty"]JPav: Thanks for this fun, well-thought out article.
I'm not sure that one can conclude that run prevention is more important than run scoring for 2007 based on your 18 top teams averaging 4th in pitching vs 5th in offense.[/quote:3a97672de3]
I wouldn't get hung up on 4th or 5th. Here I was just reporting an arithmetic average of finishes. You can be fifth in hitting behind four bomber teams who score more runs but who are also last in pitching (while your team is 4th in pitching, maybe behind teams who fared poorly in runs scored). The position in which you finish in runs scored and runs allowed is completely dependent on the individual leagues. The [i:3a97672de3]net[/i:3a97672de3] performance is what counts the most here, I think.
(And Just to clarify, the teams used were the teams with the best record in each of my 18 completed leagues. They're not all [i:3a97672de3]mine[/i:3a97672de3]. I may have referred to them as "my" data set. Sorry if this was confusing).
[quote:3a97672de3="teamnasty"]One spot of difference over 18 leagues is not much difference at all, and a sample of 18 leagues is large enough to be important but not so large as to be conclusive. If anything, your numbers suggest that balance is important to your league winners as opposed to imbalance; i.e., your winning teams tend to be slightly/moderately above average in both run prevention and run scoring, as opposed to high offense/avg pitching or high pitching/avg offense.[/quote:3a97672de3]
Net runs [i:3a97672de3]are[/i:3a97672de3] conclusive. Strong pitching and average hitting are clearly present. The OBP (.339 versus .331, which is so noticeable to the naked eye) is only a two percent improvement. The pitching stats with a seven percent better ERA and a five percent better WHIP than the average team is markedly better, imo. We may need [b:3a97672de3]ezampol[/b:3a97672de3] to explain this, he's much better at the technical explanations of statistics than I am.
However, I do agree about "balance" if by balance you mean the ability to produce net runs.
[quote:3a97672de3="teamnasty"]I'm not convinced that outstanding defensive ranges up the middle, as compared to other positive factors, has some extra special significance to building a winner in 2007. It is precisely because up the middle D is the conventional wisdom "cliched" way to win, to borrow your term, that those players are overvalued in autodrafts.[/quote:3a97672de3]
Add 'em all up and divide. Middle infield defense has been a separating factor ever since SOM online started. That's why it's in the formula!
In last year's article I wrote about What Not to Do in SOM 2006. It specifically showed that the worst teams in any given league averaged 30% worse on the middle infield range numbers than the best teams (in addition to other problems).
The best teams do it one way, the worst teams do something else. What more would it take to be convincing?
I'm not sure that one can conclude that run prevention is more important than run scoring for 2007 based on your 18 top teams averaging 4th in pitching vs 5th in offense.[/quote:3a97672de3]
I wouldn't get hung up on 4th or 5th. Here I was just reporting an arithmetic average of finishes. You can be fifth in hitting behind four bomber teams who score more runs but who are also last in pitching (while your team is 4th in pitching, maybe behind teams who fared poorly in runs scored). The position in which you finish in runs scored and runs allowed is completely dependent on the individual leagues. The [i:3a97672de3]net[/i:3a97672de3] performance is what counts the most here, I think.
(And Just to clarify, the teams used were the teams with the best record in each of my 18 completed leagues. They're not all [i:3a97672de3]mine[/i:3a97672de3]. I may have referred to them as "my" data set. Sorry if this was confusing).
[quote:3a97672de3="teamnasty"]One spot of difference over 18 leagues is not much difference at all, and a sample of 18 leagues is large enough to be important but not so large as to be conclusive. If anything, your numbers suggest that balance is important to your league winners as opposed to imbalance; i.e., your winning teams tend to be slightly/moderately above average in both run prevention and run scoring, as opposed to high offense/avg pitching or high pitching/avg offense.[/quote:3a97672de3]
Net runs [i:3a97672de3]are[/i:3a97672de3] conclusive. Strong pitching and average hitting are clearly present. The OBP (.339 versus .331, which is so noticeable to the naked eye) is only a two percent improvement. The pitching stats with a seven percent better ERA and a five percent better WHIP than the average team is markedly better, imo. We may need [b:3a97672de3]ezampol[/b:3a97672de3] to explain this, he's much better at the technical explanations of statistics than I am.
However, I do agree about "balance" if by balance you mean the ability to produce net runs.
[quote:3a97672de3="teamnasty"]I'm not convinced that outstanding defensive ranges up the middle, as compared to other positive factors, has some extra special significance to building a winner in 2007. It is precisely because up the middle D is the conventional wisdom "cliched" way to win, to borrow your term, that those players are overvalued in autodrafts.[/quote:3a97672de3]
Add 'em all up and divide. Middle infield defense has been a separating factor ever since SOM online started. That's why it's in the formula!
In last year's article I wrote about What Not to Do in SOM 2006. It specifically showed that the worst teams in any given league averaged 30% worse on the middle infield range numbers than the best teams (in addition to other problems).
The best teams do it one way, the worst teams do something else. What more would it take to be convincing?
- J-Pav
- Posts: 55
- Joined: Tue Jul 03, 2012 2:34 pm
by the splinter » Fri Oct 26, 2007 7:56 am
[quote:c760ef1c18]I'm not convinced that outstanding defensive ranges up the middle, as compared to other positive factors, has some extra special significance to building a winner in 2007[/quote:c760ef1c18]
I can't and don't try to make the game too complicated but anyone who feels that Omar Vizqel, the single greatest D card Strat has ever produced, is over valued in 07 can let me draft him with the #1 pick every time.
I can't and don't try to make the game too complicated but anyone who feels that Omar Vizqel, the single greatest D card Strat has ever produced, is over valued in 07 can let me draft him with the #1 pick every time.
- the splinter
- Posts: 55
- Joined: Tue Jul 03, 2012 2:34 pm
Return to Strat-O-Matic Baseball Online 20xx
Who is online
Users browsing this forum: No registered users and 2 guests