| Grand Slam Baseball View the Forum Registry |
| |
| 0 |
Subject: When is the New Sim coming? The Commish says... Posted by: Chris - [25540221] Thu, Aug 03, 23:44 I e-mailed the Commish with regards to when the "New Sim" will be implemented. For those of you that don't know, the New Sim will have 2 main features: 1 - HR rating is not totally worthless anymore. This doesn't mean that there will be MORE homers, just that the guys with 450+ ratings will hit mroe and the guys that are in the 200's will hit less. In other words, there will be an exact correlation between HR rating and homers/AB. 2 - More consistent pitching. Otherwise known as the "Justin Thompson Rule". Justin Thompson WILL NOT EVER have a season where his ERA is 5.50(like in L2). The great pitchers will be much more consistent. Well know more about that when we see it. Anyway...the Commish's response is that he "would guess [the new sim] would be put into the game by the end of August." There's your answer. Plan ahead... | ||||||
| 1 | TBRaiders ID: 53448220 Fri, Aug 04, 01:23 |
We could call it the Mark Clark rule for my Royals in Z2. On the flip side, my Orioles in P2 are going to get much worse with a new Sim. I wonder if they can still get away with scuffing the ball.... | |||||
| 2 | Bean Dip ID: 24456316 Fri, Aug 04, 19:18 |
I guess the next question is, what does everybody think of the new features. I guess i will start things off with a little rant of my own. My best guess is that the HR ranking was not totally meaningless before. My thinking is that the problem probably didn't lie with the sim, but in the length of the seasons. There would be far less consistency in the homerun leaders in real baseball if the season was less than half of what it is now. I think we could probably see more of a correlation between HR ranking and actual HRs if we look at career stats. That said, it might not be a bad thing to add some consistency to each individual season's stats. The pitching change i completely agree with. However, that may be more of a personal matter seeing as how I have Justin Thompson and Mark Clark on my team, and i would be pretty upset if one of them put up an ERA over 5.0. What does everybody else think? Bean Dip | |||||
| 3 | Chris ID: 44443318 Fri, Aug 04, 21:33 | sorry BD...it's a good theory, but the HR rating is out of whack. The career HR leader has 420, 2nd is 340, 3rd is 400. The doubles seem to be consistent from year to year despite the short season. The HR rating has been reviewed by the Commish, much the same way as I'm doing, and he found that the HR rating is worthless(my word, not his). | |||||
| 4 | Baldwin ID: 25440222 Fri, Aug 04, 21:34 |
I just beat a guy for a league championship who had an 80+ mil payroll and pitchers with a bunch of 500's in their stats. His Bohanon had a losing season on a team that led both leagues in wins. This is a guy with 500 control, 500 changeup and fastball, break maybe 467-490, I forget exactly. Obviously there is no way in the world this guy should have stunk up the year not to mention throwing the last game of the year for that team. I for one am glad to hear the players performances are going to be more predictible. Even if it costs me next years playoffs. | |||||
| 5 | Bean Dip ID: 24456316 Fri, Aug 04, 21:54 |
At least in my league (O2) the career stats make a lot more sense. I wonder how the HR rating is intended to translate to homeruns hit. It seems that it should be set up so that HR rating is proportional in some way to homerun/hit (ie. higher HR rating translates to more HR/hit). If this were true, a higher AVG would translate to more homeruns, simply b/c there would be more chances for homeruns. Chris, do the league leaders with low HR ranking have a high AVG ranking? If the relationship between HR rating and homeruns hit is proportional in some way similar to this, I'm not sure that it makes sense to change things. I guess the question is do we want a 250AVG-500HR guy to hit more homeruns than a 450AVG-450HR guy? I'm not sure what the consensus would be. I don't want the sim creating hits for the 250 AVG guy just so he can have more homeruns than the 450 AVG guy. Just my opinion. Bean Dip | |||||
| 6 | TBRaiders ID: 53448220 Fri, Aug 04, 22:09 |
I look forward to more predictability with stats. That being said, any player can have a crappy game and even a down season. I was reading several of the oldtimers at Grandslam whine on their message board about Gurupies. I suppose they have the current players that perform for them and would hate to see it change. I would probably continue to play this game even with it's unpredictabilty (Mark Cark in Z2 has an ERA of about 6.15 this year, kinda harsh for over 11 million a year) I think it adds to the game and takes much of the luck factor out of it. Positive changes are always good. I'd love to experiment with lots of changes. How about hiring managers? You can have the kicked back confident manager or the kickin' up dirt raising hell manager....lol. For a real change, I would like to see pitchers get pulled in the middle of an inning. I can't count the times I have been winning a game until my starter flamed in the 7th or 8th with a 5+ run inning. Or even sending in a RP and he doesn't have his stuff going, let's get back into that bullpen. As far as home run stat, I doubt it makes much difference. I will continue to build my team around speed and average. | |||||
| 7 | Bean Dip ID: 24456316 Fri, Aug 04, 22:14 | TBR--According to the commish, mid-inning pitching changes are on there way. | |||||
| 8 | 30something deac ID: 44447319 Sat, Aug 05, 10:57 | Speaking of changes, one I would like to see implemented would be the capability of disabling a player(s) from attempting to steal. I can't count the times that I've been down by one run in the latter part of a game, had one of my slower players lead off the inning by singling and consequently get thrown out (with no outs) trying to steal second. | |||||
| 9 | TBRaiders ID: 53448220 Sat, Aug 05, 17:50 |
It would be a cool change to set player aggressiveness. If any of you play Webl you know how easy it would be to set conditions for different game scenarios. inning 5) if score <= -2 agg+6 inning 9) if score <= -1 and runners in scoring position PH1 for AVG < 450 ( weird example, but I am in weird mood) That would be cool too, have Pinch Hitters set up for each game. Designated as PH1, PH2, etc. That way you could always make sure players get in game. I bet we could come up with dozens of good ideas. I know several teams are succesful always playing there best players every game. I try to get all of mine in at least two games a week with a couple studs playing all 4. ps. WeBL is a very cool boxing sim. Check it out at http://www.vivi.com/ | |||||
| 10 | Chris ID: 44443318 Sat, Aug 05, 19:28 |
BD...I'll answer your questions as best I can. - HR rating will be proportional to homers/AB, exactly like GAP rating works now(2B+3B/AB) - High HR guys will NOT hit for a higher BA. There singles will be "converted" to homers by the HR rating. AVG will still directly correlate to BA, and GAP will still directly correlate to extra base hits. A 350-500-500 guy will likely hit .200 with 12-15 doubles/triples and 12-15 homers or something like that. - Therefore, the 500HR guy will hit more homeruns than a 450 HR guy, regardless of average, much the same way that a doubles leader can have low AVG. -My "guess" is doubles/GAP and homers/HR will be the exact same formula. So, the league leaders in homers and doubles should have roughly the same number... - All of this has not been implemented yet. It was intended to work this way, but didn't work(ie shoddy programming). That's why they are changing this in the "New Sim"... | |||||
| 11 | Bean Dip ID: 24456316 Sun, Aug 06, 00:10 |
Chris--Here is why i do not believe the "new sim" will eliminate statistical disparities: Essentially, my understanding of the way the HR rankings translate to actual homeruns is as follows. Everytime a player gets a hit, there is a chance the hit will be "converted" (as you say) into a HR. How much of a chance there is for any particular hit to be "converted" into a homerun is determined by the homerun ranking. Thus, a player with a 500 HR rating may have one out of every 5 hits "converted" into a homerun, whereas a player with a 300 HR rating may have one out of every 15 hits "converted" (I have no clue what the real numbers are, but I will use these just for arguement's sake. The real numbers obviously could be very different). Thus, assuming the same number of hits (chances), the 500 HR guy should hit 3X as many homeruns as the 300 HR guy. However, in the game, players will not have the same AVG rankings. Let's suppose the same probabilities are used for translating AVG ranking into hits. Thus, a 500 AVG guy would have 3X as many hits as a 300 AVG guy. With this in mind, consider two players: Player A has 500 AVG, 300 HR; Player B has 300 AVG, 500 HR. Player B should average one homerun every 10 hits. Player A should average one homerun every 30 hits. However, player A should average 3 times as many hits as player B. Thus, since player B has only one-third the chances for homeruns as player A, they ought to end up with the same number of homeruns over the course of a season. This is my understanding of how the "old-sim" works. Managers obviously had a problem with this system, as many times it would leave guys with great HR rankings with few actual HRs. I can only think of two ways to make players with low AVG and high HR rankings hit more homeruns. The first is to simply increase the number of homeruns/hit that these players are programmed to average. However, this would not fix anything. A 300 AVG 500 HR guy will always (over the long term) end up hitting fewer homeruns than a 500 AVG 500 HR guy, simply because the 500-500 guy will have more chances. Increasing the homerun/hit ratio would just make everyone hit more homeruns. The other way to increase the number of homeruns a low AVG, high HR guy will hit is to simply have the computer "create" more HRs for such a player. However, while this would fix the disparities in homerun stats, it would cause equally egregious disparities in batting average stats. Given that a homerun will always count as a hit, this would just cause players with low AVG and high HR rankings to hit for unreasoably high batting averages. I don't want a 250-500 guy hitting for the same AVG as a 450-300 guy, just because the computer "created" 10 homeruns for him, so he could have the same number of homeruns as a 500-500 guy. It seems to me that as long as homeruns count as hits, the HR and AVG rankings will always affect eachother in one way or another. Disparities are unavoidable. All we are doing now is choosing which stat we want to be unfairly skewed. If the concensus is that we would rather have screwy averages, so be it. However, this does create some problems for managers who have been drafting already taking into consideration the present idiosyncracies of the sim. But then again(as Dennis Miller might say), that's just my opinion, i could be wrong. Bean Dip | |||||
| 13 | Chris ID: 44443318 Sun, Aug 06, 01:02 |
Let me try to clear this up. Mind you, all of these examples are how the new sim will work... AVG is hits/AB, GAP is (2B + 3B)/AB, HR is homers/AB. This is one way it could work, and believe me, it can work as there is no discrepancy in GAP, and HR will be worked EXACTLY the same way. One player has 500AVG, 400 HR. Another has 400 AVG, 500 HR. The first player has a .35 probability of getting a hit and a .032 probability of getting a homer. The second player has a .25 probability of getting a hit and a .056 prbability of getting a homer. Now, how can we work a formula to make this happen? First off, run a random number generator to see if he gets a hit. If you do it this way, their AVG's will be correct, right? Anyway, if he does get a hit, run another random number generator. The first player has a .032/.35 chance of getting a homer. The second player has a .056/.25 chance of getting a homer. This way, the homers will be correct. Does that make sense? | |||||
| 14 | Chris ID: 44443318 Sun, Aug 06, 01:06 |
Let me explain where I got the numbers... .35 = .350 batting average .25 = .250 batting average .032 = 8 homers in 250 at-bats, what I might expect a 400 HR guy to do in a full season(roughly 250 AB's) .056 = 14 homers in 250 at-bats HTH | |||||
| 15 | Bean Dip ID: 24456316 Sun, Aug 06, 11:41 |
Chris--I still saw problems with any system where another random generator is run after each hit. Any such system would result in players with more hits (higher AVG ranking) getting more homeruns, simply because they have more chances. Because of this, i emailed the commish. He said a couple interesting things in his response. First, he gave me an idea of what stats the rankings are supposed to correlate to: "Over the course of 100 at bats, here's what we would expect: 500 AVG, 500 HR = 35 hits, 10 of them HRs. 500 AVG, 300 HR = 35 hits, 4 HRs. 300 AVG, 500 HR = 20 hits, 10 HRs. 300 AVG, 300 HR = 20 hits, 4 HRs." I had never seen before what actual statistics were supposed to be produced (on average) by hitters with certain rankings (although we certainly had guessed). This does prove your -150 theory right Chris. Also, my best guess is that doubles work the same way, in the same proportions, as homeruns, and that players with great speed get a few triples out of their doubles. He then went on to say the following of how disparities in batting average would be accounted for and corrected: "If you assume that all of these players have the same GAP rating then they will have the same number of 2Bs and 3Bs. Then you will see that the number of singles each player hits will adjust so that both their batting average and homerun total is accurate." Given this, i think the way he is planning to do it is to "create" homeruns, as i thought he might, for players with low average and high HR. However, then to correct the situation, he will somehow "subtract" hits from these players so that there averages will not be out of whack. Oh, and by the way Chris, you are right about HR rating corresponding to HR/AB and not HR/hits. Well, at last some substantiated answers. Plus, i think it is really cool that he just threw in what the rankings were supposed to correlate to in statistics. This should save you some work in doing all those tables Chris. Thank you for all the effort, and for getting us started on this little quest for understanding. I hope all this has given everyone some good food for thought. Hopefully, a lot of people will be able to build better teams, using this info as a foundation. Bean Dip | |||||
| 16 | LTA ID: 18448222 Sun, Aug 06, 12:29 | Hmm. Well, so much for the couple of hours of data-mining that I had put in. *Poof* Question answered. ;-) | |||||
| 17 | Chris ID: 44443318 Sun, Aug 06, 18:02 |
BD...I'm not saying that IS the system they will use, I was only trying to say that it can be done. All the Commish did really was confirm everything I've said up until this point. There is a point in collecting data still, however. It'll be nice to know where the middle ground lies(ie how many homers will a 460 guy hit?). That data is still very useful LTA...thanks. I don't think he is "creating" homers and somehow subtracting hits. I still believe that singles are "turned into" extra base hits. I'm not sure exactly how it is coded, but I'm sure this is how it's done...Remember, he can't subtract hits because each AB is a separate event. There has to be one formula he runs for every AB... | |||||
| 18 | Chris ID: 44443318 Sun, Aug 06, 18:12 |
Let me use my previous example... One guy has a .35 chance of getting a hit and a .032 chance of getting a homer. The other guy has a .25 chance of getting a hit and a .056 chance of getting a homer. Let's assume that each player has a 0% chance of getting a double or a homer. For both players, a random number generator is run. For the first player, there is a .318(.35-.032) chance of getting a single, and a .032 chance of getting a homer. For the second player, there is a .194(.25-.056) chance of getting a single and a .056 chance of getting a homer. In this way, each stat can be kept separate of the others. This is the way I would do it... | |||||
| 19 | Chris ID: 44443318 Sun, Aug 06, 18:19 |
BD - On the GSB Message Forum, the Commish responded to my first post, which I copied and pasted from above. Here is what was said... "It makes sense to me Chris, because this is very similar to how it really works. The HR rating determines HR/AB, not HR/hit. So while it's true that you have to be able to get a hit first before it's "turned" into extra bases (as Kurt posted), the AVG rating does not effect the number of extra base hits. This is because the lower the AVG rating, the higher the percentage of hits that go for extra bases. If two players have the same GAP and HR ratings, then they will have the same number of extra base hits, but the guy with the higher AVG rating will have more singles making his batting average higher." Which is pretty much what I've been saying all along... | |||||
| 20 | Bean Dip ID: 24456316 Sun, Aug 06, 19:19 |
Chris: First of all, i want to say that i agree with you on almost all of this (and have all along). The disagreement here has been over a very small (but still significant, i think) point. I think "create" and "subtract" may have been the wrong words to describe what i meant. Let me elaborate. I think the point can be best demonstrated using a player with very extreme rankings. Assume a player has an AVG ranking of 1 and a HR ranking of 500. Let's say the player gets 100 ABs in a season. He should, according to the commish, average 10 homeruns per 100 ABs, as this is the model for a player with his HR ranking. Unfortunately, there is a problem. Supposedly, homeruns are hits that get "converted." The greater a players' HR ranking, the greater the probability of a hit being "converted" into a homerun. However, this player, even if he gets really hot, will have no more than 3 or 4 hits in his 100 ABs. A discrepency here is unavoidable. One possibility is that he will have just a few hits, and accordingly just a few (at most) homeruns. This will mean that he is hitting for an appropriate AVG, but not coming anywhere close to hitting as many homeruns as he "should." The other possibility is that the sim will somehow "create" (i can't think of a better word) homeruns for this player so that he can live up to his rating. Seeing as how each homerun counts as a hit, he will end up hitting for an unreasonably high average. Now, obviously our ficticious player is an extreme example. Almost all players with such a high HR ranking will have a AVG ranking over 300. Nonetheless, the discrepency will exist. The only way to get around this problem is to make the percentage of single "converted" into homeruns higher for players with lower AVG rankings but equal HR rankings. Thus, a player might end up having 75% or more of his singles be converted into homeruns, just to satisfy both HR and AVG rankings. I believe that this is what the commish intends to do. Through this kind of tinkering, it will be possible to make the HR and AVG stats appear independant. The commish is correct in saying that the problem will be fixed. However, i believe that the 1AVG-500HR player demonstrates how it is not fully possible to have the correlations between HR-homeruns/AVG-batting average be totally free from eachother, and at the same time always correct. Bean Dip | |||||
| 21 | Chris ID: 25540221 Sun, Aug 06, 20:32 |
"The only way to get around this problem is to make the percentage of single "converted" into homeruns higher for players with lower AVG rankings but equal HR rankings. Thus, a player might end up having 75% or more of his singles be converted into homeruns, just to satisfy both HR and AVG rankings. I believe that this is what the commish intends to do. Through this kind of tinkering, it will be possible to make the HR and AVG stats appear independant. The commish is correct in saying that the problem will be fixed." That is what I've been trying to say. I don't think we actually disagree. The algorithm above(post #18) does exactly this. It's even easier to see this in my other algorithm(post #13). In that one, after the guy gets a hit, the percentage is different for each player. Let me copy and paste... "Anyway, if he does get a hit, run another random number generator. The first player has a .032/.35 chance of getting a homer. The second player has a .056/.25 chance of getting a homer." As you can see, this is exactly what you and I think will happen. I also think that this is the way they currently worrk the gap rating. The 2nd algorithm uses the same principles, but is coded much better. I think we're all on the same page... As far as the 1AVG example. I'm sure there is somewhere in the algorithm that states a guy must hit at least .100 or something. If you stick to the -150 rule, a player with 1 AVG will have a negative BA, obviously this can't happen. I believe that the formula is one-to-one linear down to a certain point . Once it gets to that level, there is a different formula, still linear, but not 1-to-1. The slope of the line changes at that point... Does this make sense at all? | |||||
| 22 | Phat Cat ID: 24456316 Sun, Aug 06, 21:19 |
I was assuming that the -150 rule didn't hold all the way down. In fact, given that the commish said that a 300 AVG corresponds to a .200 average, i think there is a pretty good chance that the difference between AVG ranking and intended batting average is constantly decreasing. This would make a person with a 1 ranking have roughly a .000-.001 batting average. But anyways, you understand. I'm not sure what we will do now that there is nothing to argue about ;-) It was fun while it lasted. Bean Dip | |||||
| 23 | Chris ID: 25540221 Sun, Aug 06, 22:37 |
lol...There is ALWAYS something to argue about in my experience... ;-) Be patient, as I'm collecting GAP data now...I should have the bulk of it by Tuesday. Anyway, it may be constantly decreasing, although 400 guys still look like they hit .250 in my experience. That MAY be the cutoff point for which the slope of the graph changes. If a 300 guy should hit .200, maybe the formula has been halved, so that... BA = AVG/2 + 50, where AVG is less than 400. Just food for thought... | |||||
| 24 | Chris ID: 25540221 Sun, Aug 06, 22:43 | ...and it's a good thing you always attach your name to your post...otherwise I would have no clue who I was talking to ;-) | |||||
| 25 | Bean Dip ID: 24456316 Mon, Aug 07, 01:14 |
Just looking out for ya. Bean Dip | |||||
| 26 | Chris ID: 235563022 Mon, Aug 07, 01:21 |
thanks Chris | |||||
| 27 | M ID: 527840 Mon, Aug 07, 22:02 |
Personally, I kind of like how the pitchers are somewhat inconsistent. I've had Justin Thompson for four years and he has put up a 2.45, 2.78, 3.10, 4.10. I believe this is very realistic. Besides Pedro and Randy, no pitchers have dominating years every year. I think it keeps you on your toes and forces you to build depth within your team. I think the home run thing is out of whack, but the real problem is not enough homers are hit. Guys with 320 hr end up winning the home run race because they just got hot. Two home runs in a series is a milestone! I've been in two seasons where the top home run getter had 7 home runs! 7! Maybe someone had three good series and hit five over. All he has to do is hit four more in the last 13 series and boom, hes the home run king | |||||
| 28 | Chris ID: 235563022 Mon, Aug 07, 23:48 |
Maddux, Randy, Brown, Pedro, Smoltz. These are the Thompson's, Clark's, and Bohanon's. These guys don't have ERA's of 5.50...Also, if we are going to have entire teams hit .320, like it is now, we can't increase the number of homers. It would be like beer league softball... | |||||
| 29 | Jaysuade ID: 3449412 Tue, Aug 08, 01:34 | There has been a change to the new sim that has lowered batting averages.I have a macro for excel set up to get the data for HR and AVG but I am waiting for the season to end in the test league which should be this Wednesday.These macros only work for the test league.I have tried to figure out how to get them to work for the regular leagues and IMO I don't think it is feasible with the way they are set up now. | |||||
| 30 | blue hen ID: 386331314 Tue, Aug 08, 14:36 |
I once wrote a program that determined outcomes of at-bats. Actually, I wrote a program that played a whole baseball game, but the outcome logic was the most important part. I took a season and prorated it to 1000 at-bats. Let's consider "Sidd Finch". 500 at-bats, 143 hits for a .286 bat average. 25 of those hits were homers, .050 homer avg. 33 of those hits were doubles, .066 double avg. 8 of those hits were triples, .016 triple avg. 77 of those hits were singles (do that math), .154 single average. Then run the sim. Enter a random number between 000 and 999. 000-153 single 154-219 double 220-235 triple 236-285 homer 286-999 out That's your hits. Then you can divide the outs into parts. It was easy logic. I'll venture to say that the lower "out" number is set by AVG rating minus 150. 2B/3B and homers are added accordingly, and the rest are singles. Do you have a 200/500/500 guy? I'll bet the 500's do indeed "raise" the number at the bottom of his range of outs. I didn't go that in depth into this, but most of that is because I simply will put my faith in the hands of the Commish. | |||||
| 31 | Chris ID: 235563022 Tue, Aug 08, 17:39 | That looks an awful lot like my 2nd algorithm... | |||||
| 32 | Swoosh ID: 375203020 Fri, Aug 11, 17:28 | Does anyone know what will be the most important stat in pitching in the new sim? contol, breaking, change? And what numbers do the sub 4 era pitchers have? say above 400 in each? or a total of all 4 categories in the 1550's? like 400 fastball, 300 change, 450 change, 400 control, 400+300+450+400=1550... Thanks for the help! | |||||
| 33 | Timing ID: 67132818 Wed, Aug 30, 17:03 | Well, it's the end of August and there isn't a new sim in sight. Sup with that? | |||||
| |||||||
Post a reply to this message: (But first, how about checking out this sponsor?) | |||||||