| RotoGuru Baseball Forum View the Forum Registry |
| |
| 0 |
Subject: Smallworld Pitching Statistics, Part I Posted by: Madman - [146191423] Tue, Feb 20, 17:34 In a few threads, two good topics have been brought to the fore. First, what about ballpark effects as they affect SWP? Secondly, what are the differences in the point scoring formulas, and do these differences have any implication for strategy? I'm going to recreate an infamous table from the 2000 SW season. What you'll find below are a wealth of pitching statistics for starting pitchers only in the 2000 baseball season. All points have been converted to 2001 SW Points. I'll give a brief analysis of various components of the table in a follow-up post. But first, some definitions are in order. Some Definitions "Ballpark" information represents a) the number of starts that pitchers threw in a ballpark, b) the number of points they would have scored with the 2001 SW scoring system, and c) the average number of points per start. These columns are in green. The purple columns, under "Team" represent the pitching staffs of the respective teams. I think the stats are self-explanatory. The blue points-against columns represent how well opposing pitchers fared against the team on a given row. For example, starters against Anaheim would have scored an average of 24.9 points with the current scoring system. Finally, the "Road" columns (yellow) represent how well pitchers fared in games in which the given team was on the road. For example, when Anaheim travels to KC, the stats from BOTH starting pitchers in those games would count in the ROAD totals for ANA and in the HOME totals for KC. Finally, I calculated a very simple "ball-park factor" in the last column. This is just the difference in the home-road games. A positive number represents a pitcher's park, a negative one represents a hitter's park. Note, however, that this information relies solely on starting pitchers.
Notes: *Cin and Milwaukee played in 163 games last season because of the tie on opening day. *Statistics were compiled primarily from ESPN boxscores, with some data coming from other sources. Post-game scoring changes may not be reflected in this data. Errors are the sole responsibility of the author, and not ESPN. to be continued . . . | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 1 | Madman ID: 146191423 Tue, Feb 20, 17:51 |
1) If you were to examine the same statistics using the 2000 SW point formula, you'd find that the average start accumulated 38.3 points. The new formula drops that to 29.9. This is a 22% reduction in points (on average). This will surely impact your pitching v. hitting strategy. 2) The best parks to pick up SW Pitchers -- all else equal? SD, CHI (Wrigley?!?), SF, LOS, SEA, with OAK not far behind. The worst? Coors (duh), CLE, Ari, KC, CHI (WS). 3) Notice that the presence of Coors field makes a large number of parks look like pitcher's parks. It's probably not fair to compare NL and AL ball park factors with this method (which, BTW, mirrors most methods for computing ballpark factors -- namely comparing the road/home performances of teams). 4) Also notice that you shouldn't look at Ball-park factors alone (duh). Tampa's -5.9 ballpark factor was vastly outweighed by their rather inept offense last season. Pitchers against Tampa Bay scored a whopping 42.2 points, or 41% higher than a team-neutral, ball-park neutral performance. -------------------------- The most important things to learn from this chart? a) Low point totals. Pitcher point performances are going to be rather low this year, b) Match-ups. The match-ups are going to be exaggerated in importance with the new scoring formulas. There is a 20-25 point gap between the expectation of a pitcher facing a good offense versus a bad offense. This difference is almost as big as the value of a generic start! c) Patient Offenses. Notice also that "patient" offenses tend to really hurt opposing pitchers under this new scoring system. Patient offenses always are tough to face, since pitchers tend to leave the game with fewer IP's against them. Now, however, there is an increased penalty for BB, which really makes teams like OAK and such dangerous. Even SEA was very dangerous with this point-scoring formula (note: past returns are not necessarily indicative of future performances, of course). --------------------------------- Next, I'm going to examine some additional statistics from the 2000 season to examine exactly how the new scoring formulas may affect strategy. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 2 | JKaye ID: 4711592917 Tue, Feb 20, 17:52 | Madman--2 of the best threads of the year and it is only March! Great job. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 3 | Strike One ID: 271552014 Tue, Feb 20, 18:04 | it seems that a strategy where one would load up on hitters this year would benefit due to the changes in the scoring system. i think it would be interesting to look at pitchers individualy as well. which pitchers are gonna be more effected by the scoring change by smallworld? if only i knew how to compute all the info to come out on a spreadsheet...... | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 4 | Madman ID: 146191423 Tue, Feb 20, 18:08 |
Not even March :-). Actually, however, this was pretty easy to do. I did all the work last season. I just punched in the new formula, and out pops all this cool information :-). Correction I made a mistake when explaining why Patient Offenses are more dangerous now. If you look at the point formula, a BB costs exactly the same as it did before. The only significant difference in the formula is a reduction of the benefit of a K, from 5 to 3. This accounts for the vast majority of the reduction in points-scored (note: pick-offs account for the remainder). But this impact is HUGE -- especially for teams like OAK and SEA which were very patient. Their patience resulted in many walks, but it also made them susceptible to K's. Now, it takes 1 and 2/3 K's to compensate for each BB instead of the 1 to 1 ratio from before. Thus, high strikeout teams are more dangerous to face than they were before. But I stand by my overall point that patient offenses are the ones to avoid, since they tend to K a bit more. More importantly, the fact that pitcher's tend to reach their pitch-counts earlier is critical, as well. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 5 | Strike One ID: 271552014 Tue, Feb 20, 18:48 |
Ok, i did a little research by hand here and i've come up with a list of which pitchers were most affected by the change in the strikeout forumala (5 SWP--> 3 SWP) Rank, Player, Last Year' SWP, This Year's SWP, SWP difference 1. Randy Johnson 3800 3118 694 2. Pedro Martinez 3855 3287 568 3. Chan Ho Park 2510 2076 434 4. Kevin Brown 2995 2563 432 5. Bartolo Colon 2075 1651 424 6. Mike Mussina 2295 1875 420 7. Ryan Dempster 2220 1802 418 8. Al Leiter 2425 2025 400 9. Javier Vazquez 1895 1503 392 10. Rick Ankiel 1980 1592 388 could this possibly be another stint by SW to stop the mad RANDRO? | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 6 | quietriot0 ID: 157442319 Tue, Feb 20, 19:06 |
The shift in pitching would seem to lesson the the difference between power/dominating pitchers and those who just get the job done(don't give up runs). I assume it'd hurt Randy and Pedro more than others, but maybe not much more relative to their total points. Personally, I like the change. Who cares if a pitcher didn't strike out 10 batters when he doesn't give up an earned run. Also. Another note on the scoring being 22% less this year for pitchers. You might also want to consider that the strike zone has been enlargened, which might very well negate or even reverse the decline in pitching power. It isn't really something that you can guage, but some pitchers may benefit tons from the enlarged strike zone. Maybe lots won't benefit as well(throwing high in the strike zone w/o overpowering stuff isn't really all that safe). So if you are thinking about going cheaper w/ pitchers since they will be 22% effective(even though i doubt much anyone would), you should consider the change in the strike zone as well. It'd be interesting to know who will benefit the most from the change in the strike zone. Someone linked to a Peter Gammon's article that has statistics on the best and worst fly ball and ground ball hitters and pitchers. Of course, these statistics probably won't translate directly and completely to stat improvements/decreases(as the article does note). | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 7 | Madman ID: 146191423 Tue, Feb 20, 19:12 |
OK. Now to the issue of the differences in the pitching formula. First, consider the following graph that examines the SW Points scored by STARTING PITCHERS in the 2000 season: The above chart shows the relative frequencies of various outcomes, broken down into categories of 15 SWP each. The table should be read as follows: The first two columns show the probability that any start will score -106 or worse. The second pair of colulmns show the probability that any start will score between 106 and 90, etc. The blue columns are for the 2000 formula, the red columns are for the 2001 formulas. Look at the relative frequencies of SWP's for all categories of scoring less than 0 points. Notice that the 2001 formula has MORE starts in this category for EACH range. Specifically, 32.0% of all 2001 formula starts fell in this range, while only 27.9% of 2000 formula starts were in the same region. Similarly, on the other end, look at starts that scored greater than 90 points. The 2001 Point formula dramatically reduces the number of such starts (the only exception being a statistical tie for starts between 120 and 135 points). Specifically, with the 2000 point scoring system, 22.3% of all starts scored more than 105 points. With the 2001 system, only 18.1% of starts fell into this range. In other words, this scoring change increases the risks of negative performances, and simultaneously decreases the chance of exceptional ones. Specifically, you're going to run about a 4% higher chance of a negative point outing, and a 4% lower chance of a 90+ point outing. This is ignoring the obvious influence that you can have on these statistics by choosing good quality or bad quality pitchers. This will have multiple effects. First, the negative psychological impact of negative starts might further exacerbate the flight toward better quality starters and c-losers. Secondly, it will make it more difficult to obtain pitching points. Thirdly, the incredibly large positive point values will be much fewer and farther between. This will close the gap between Pedro, Randy and the rest of the field. Fourth, the standard deviation of starts is slightly lower, meaning that any given start has slightly less variation. That's about the only bit of good news I can glean out of all this :-). Of course, keep in mind that if they did the SW pricing formula correctly, the observation about the relative value of a strikeout pitcher is already incorporated into the actual prices of pitchers (checking this is on the to-do list). | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 8 | Los Loco Pollos ID: 91492114 Wed, Feb 21, 19:03 | madman you are a genius! what program are you using to create all thoes cool graphs and stuff? | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 9 | GoatLocker ID: 23727611 Wed, Feb 21, 19:14 |
I would be willing to bet that the graph came from Excel. Great work Madman. Thanks, Cliff | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 10 | Madman ID: 146191423 Wed, Feb 21, 23:29 |
Yep -- Simple Excel 2000, then published the chart as an HTML file. The chart was created by using the FREQUENCY function -- I set up a grid of "bins" from -106 to 195 in increments of 15. I also have an add-in called Stat-Pro that one could have used to facilitate the whole process . . . I downloaded all the starts as part of a rather complicated macro/spreadsheet program I created last April, and tinkered on all season. I was going to release it to the public, but I got some bug or something that caused my Excel 2000 to take FOREVER to run the macro (you'll just have to trust me when I say it was complicated). I then installed and converted it back to Excel 97. However, it was always a bit "buggy" after that, and I didn't want to release such a product publicly. Unfortunately, my Excel 2000 is still slow for processing the code. This really kills my ability to troubleshoot large and complicated spreadsheets. I've debated about converting the whole thing into VB for the 2001 season, but that will take too much development time. . . Sorry, I'm not a professional programmer :-( | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 11 | Eustacio ID: 51729258 Thu, Feb 22, 06:35 | Madman, what's the problem exactly, running all of the mathmatical computations or downloading the information via the macro? | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 12 | Madman ID: 146191423 Thu, Feb 22, 15:20 |
Eustacio -- I have approximately a dozen different macros of significant length in the whole spreadsheet. One known problem is that the file size is too big, and it takes several minutes to boot on my P-III 500. I've thought a lot about this issue, and there's not a lot I can do about it. I have almost no equations to be calculated live-time on the spreadsheet (to speed up the program) -- I use macros and subroutines almost exclusively, so nothing gets calculated until I need it in a macro. As I said, this is a very complicated piece of work. The only time I use spreadsheet equations are to do a quick analysis of something. I really should be running this in Microsoft Access, although there are a couple of difficulties I'd have in that format . . . More troubling, however, is the macro issue, which seems to be related directly to a file-opening problem, as well. Primarily, one of the macros goes through a schedule of games and downloads all of today's games (if they occurred). The problem is that opening a single HTML page from ESPN seems to take on the order of 3-4 minutes. It then takes forever to close the file afterward. Note: I have disabled screenupdating and auto-calculation, so the user never sees it being opened or closed. Just this super-long pause . . . For a 15 game-day, this means that my macro would daily take 45 minutes to run!! If I just open the ESPN pages without my huge spreadsheet in the background, I'm still getting slow load times. (i.e., I'll even save the HTML file to disk, and then "open" and the thing takes literally forever with nothing else going on). I must have reset some HTML option somewhere along the way, but it's impossible to find. To make things especially weird, certain HTML pages load almost instantly when I do this procedure. I think it *might* have something to do with the presence of control elements or forms on the HTML page . . . So, actually, the problem is most definitely not in my programming (directly), but in my version of Excel. I've uninstalled, re-installed, re-installed partially, you name it. . . A solution would elevate the provider to psuedo-god status in my eyes . . . | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 13 | miguel p ID: 7157140 Fri, Feb 23, 16:02 | Madman -- thanks for all the effort. I have a quick general question. I'm new at this, so I don't know what the scoring system was last year, but I'm curious how much the new scoring system subtracts from the old hitters' swps (as an average percentage). If it's less than 22%, then both this and the fact that the standard deviation for (starting) pitchers' swp/g is lower suggest that one should decrease their $ spent on pitching / $ spent on hitting ratio from what it might have been based on last year's numbers. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 14 | Madman ID: 146191423 Fri, Feb 23, 16:18 |
miguel p I haven't had time to sift through the hitters yet. What you discuss is important. An "eyeballing" of the scoring system for hitters would indicate that power hitters are going to be less productive and that speed guys are going to be helped. Some GUESSES: a) In aggregate, offense is slightly reduced -- definitely reduced by less than 22%. b) However, offense for the stud players -- the "All-Stars" you lock in -- is probably reduced by more than the average hitter. But still not 22%. c) Regarding the $$$ on pitching for $$$ on hitting, there would need to be one more step in the analysis -- to what extent are the current SW prices "efficient"? If they perfectly adjusted the prices of hitters and pitchers, then you're right -- spend more money on hitting than you would have last year. The complication is that there are "bargains" out there. And these "bargains" are going to form the basis for your team. The ultimate % of money spent between pitching and hitting will likely reflect the relative scarcity of these bargain players as much as it reflects underlying productivity issues between offense and pitching. In general, however, the scarcity/efficiency arguments were present last year. Therefore, I would tend to agree with you that it's likely that relative to last year, you may want to pump a bit more money into offense . . . Actually, you may have no real choice in the matter, since the price of pitchers is likely reduced by more than the prices for hitters -- as per the arguments expressed above. One last huge caveat: $50m goes a lot further than it did last year. Specifically, it's probably worth $70m in 2000 draft-day prices. With that extra cash, it would be reasonable to buy some mid-ranged pitchers to reduce your variability . . . In other words, in the short run, you may still pump money into pitchers, but in the long run, you're right -- hitters are going to be where the points and money eventually flows . . . relative to last year . . . I haven't checked any of this post out -- all pure speculation so errors in logic, etc., are quite plausible. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 15 | Wammie ID: 20039259 Fri, Feb 23, 16:27 |
Early people were talking about Wrigley getting the high pitcher's park rating. Even though the cubs hitting last year was bad, their pitching ERA was crap. so i would not blame the hitting. It just shows that Wrigley is not the "small park" that it gets the rep for. When the wind blows in at wrigley can't be a park that would better to pitch in. But if the wind is blowing out, the bat boy could probably go long. Wrigley was one field I never consider as a hitters park or pitchers park during the season. It could go either way. but in the spring and fall, the wind typically blows in, and in the summer, it blows out. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 16 | miguel p ID: 7157140 Fri, Feb 23, 16:38 |
Madman, who the hell do you think you are, making me wait 16 minutes for a response? : ) Thanks again. You rule. The only other thing I was thinking was in terms of the relative standard deviations of average hitter swp/g and average pitcher swp/g. If the standard deviation for pitchers were significantly lower, that might compel me to stay away from the better pitchers (those in the higher percentiles in terms of swp/g), knowing that they would not produce as many more points relative to the average pitcher as the better hitters would produce relative to the average hitters. Of course, this all depends on the pricing algorithm SW used to set the prices -- specifically, on the exponent in some equation "Price = (some measure of sw-production)^X" which accurately approximates the relationship between Price and Production over some interval of Production. I would expect X to be greater than 1, at least once you get to the average player range (the marginal product of any dollar spent should be decreasing at that point, IMO), but I don't know if we can trust SW to use the expected / most sensible system. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 17 | Sludge ID: 1440310 Fri, Feb 23, 16:44 | Careful with them formulas there, miguel. Being a thread killer is an exclusive club. We may have to let you in. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 18 | Madman ID: 146191423 Fri, Feb 23, 16:49 |
miguel p I just happened to be around :-). My graph in post 7 represents all start, not starters. I haven't looked to analyze the distribution of starters and/or their standard deviation. They probaby have shrunk as well, since the scoring formula only changed on one dimension. Overall, however, I'm not sure that this reduction in the standard deviation is really all that significant. The elite pitchers may have come back to the pack a bit, but my guess is that their price reductions (i.e., our increased effective wealth) have more than compensated for their slightly lowered production. And your points about the SW price formula are critical. I honestly haven't looked at how the draft day prices were set. That's on the agenda, as well groan. --------------- Wammie I totally agree about Wrigley. You're prey to the weather. If this cold winter in the Midwest carries over to April, be advised that Wrigley could still be a pitcher's park early in the season. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 19 | quietriot0 ID: 157442319 Fri, Feb 23, 18:48 |
I was looking at sw price settings about a week ago. I have all the data on my other computer, but as I remember it price is a linear graph based on total swp's per season. The relationship seemed to be SWD(in millions) = .5 + SWP/600. There are random fluke players who are valued at much more than this. I _think_ that sw adjusted their values based upon their popularity(how many people owned them at one point) last year, but i'm not really sure. And of course there are other extreme cases such as John Smoltz where he didn't get any points but is still worth quite a bit. I'm not sure how it all works out. If I get some charts out of the data I'll post them here. The one implication I noticed is that in the weaker positions(I looked at SS) the top players were very heavily owned and therefore will get you much less bang for your buck. At positions like 1B, ownership is spread out more and players will get you significantly more bang for your buck. So it seems that you actually should draft sleepers at the weaker positions | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 20 | Madman ID: 146191423 Sat, Feb 24, 17:31 |
One major criticism of sabermetric adjustments for Ball-park factors is that they really don't take into consideration the quality of pitchers making the 162 starts in each each ballpark. They rely on a "large sample" size to determine this. But a misplaced Pedro injury could give him more starts on the road v. home. Bad luck does happen. The following chart is an attempt to take this into consideration. Rather than calculating the average number of SWP's scored in each ballpark, this chart only shows each pitcher's deviation. In other words, RJ scored -95 points in his last outing at PacBell. Rather than averaging this -95, I put in a -183. RJ normally scored +88 (average for whole season -- yes, including the PacBell start. This isn't perfect). Therefore, a hitter's park would tend to result in more NEGATIVE deviations, while a pitcher's park would result in more POSITIVE deviations. As RJ's start in PacBell illustrates, there's still a high degree of random chance involved. As you would expect, the results aren't hugely different than the chart in the first post in this thread. But if you were to hold a gun to my head and tell me to predict Ball Park factors, this is the chart I'd use. . . For ease of comparision, the last column on the right is the "old" ball park factor from above. . . If any sabermetricians are reading this, you should start to do this (and more) to make your ball-park factors more plausible. . .
(Note: Team totals are not zero for each team because of traded pitchers . . . I was freaking out forever about that . . .) | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 21 | azdbacker ID: 230212320 Sat, Feb 24, 19:02 | One of the best threads I've seen. Great work, Madman. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 22 | Pond Scum ID: 54420321 Mon, Mar 05, 16:28 | BUTT, thanks for all the effort, Madman, your threads are getting pushed out of sight too quickly. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Rate this thread: | If you wish, you may rate this thread on scale of 1-5. Ratings should indicate how valuable or interesting you believe this thread would be to other users of this forum. A '5' means that this thread is a 'must read'. A '1' means that this is a complete waste of time. If you have previously rated this thread, rating it again will delete your previous rating. If you do not want to rate this thread, but want to see how others have rated it, then click the button without entering a rating, or else click here. | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Post a reply to this message: (But first, how about checking out this sponsor?) | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||