Tuesday, May 15, 2007
3:00 PM - 5:00 PM EST
The LIVE portion of this chat will begin on
Tuesday, May 15, 2007 at 3:00 PM EST.
Welcome to our first developer chat following our monumental April 9th update. The update was HUGE. There were some initial kinks that had to be ironed out, but the end result is/was a better game with much more depth, more statistics and more control. Now's the time to ask any questions regarding the update. So, onto the questions...
What is the plan for giving us back individual base running aggressive settings? It is a big loss (assuming of course that those settings used to impact gameplay.....) (rickr - Hall of Famer - 3:11 PM)
Good question and one that's been raised quite a bit in the forums. We decided to remove the individual baserunning settings from the game for several reasons. First, there aren't any historical baserunning stats to use in the engine (bases taken versus opportunity) so the only real "stat" we can use is our speed rating which is derived from other raw stats including runs scored, stolen base attempts, etc. It's similar to the Bill James speed score. Second, it's use in the previous engine was limited. In the new engine, it's used anytime a runner has to decided whether to take the extra base. Since situation and speed dictate the decision to take the extra base as well as the success rate, it's something the manager can do without having the user adjust the settings himself. The manager can make the adjustment for the entire team, and the manager is smart enough to know to increase some guys more while leaving others be. The end result is more attempts without seeing your slowest guy on the field making way more attempts. The only thing adding individual settings back in for this would add would be the ability to make your slower runners get out more on more attempts. It's a simplification we thought was worth making.
What was the rationale for eliminating the indvidual player settings for stealing, pinch hitting and the others. It really eliminates a lot of options that we previously had to manage the team. Are you considering putting them back. (alston - Hall of Famer - 3:15 PM)
Similar question. The individual basestealing and bunt settings were left in, because there are actual stats behind these numbers. Some managers want a guy with a high SBA% and high SB% to run at full speed while not running their high SBA%, average SB% guy very much. Rather than make the manager the scape goat in his instance and having an AI-based approach like baserunning, we wanted to keep this in the hands of the manager (you guys). Pinch-running and pinch-hitting are items we believe can be coded well enough to not require individual settings. It's a simplification to the game that has continued to grow in its complexity. The PR/PH logic is not perfect yet (we continue to receive feedback and make improvements, with an update due out later today). As long as owners continue to send in boxscores for review with rationale behind the decisions, we'll continue to improve the logic. Bottom line, though, is we believe an AI manager should be smart enough to handle these decisions with an overall managerial setting and the underlying player stats and game situations.
What does a 2 setting do to pull a pitcher TPC vs MPC; what are the expectations/rules here? H & R: does infrequent mean not usually early in game and increases later in close games vs. does infrequent mean not usually when either low speed base runner or low contact hitter is present? (team_acdc - Veteran - 3:21 PM)
The call bullpen setting is a slider-type setting, much like all the other settings. By adjusting the setting below or above the average/balanced setting, you're basically widening/narrowing the range for certain events to take place. A Call Bullpen of 2 means your manager will be more patient in his decision to yank the pitcher. The # of hits/runs allowed in the game and in the inning would be more than a pitcher with a call bullpen of 3, for instance. There are many factors that go into a manger's decision to pull a pitcher. The call bullpen plays a big role in the final decision. In short, a manager will give a pitcher with a 2 call bullpen setting more leeway than a 3. The hit and run setting works the same way.
How are doubles and triples being determined now? Are hitter's stats going through a log5 eqaution? (JohnGPF - Hall of Famer - 3:28 PM)
We now estimate the # of doubles and triples a pitcher allowed following some regression analyses which proved very reliable. We now also use a weight factor for both the batter and pitcher. It used to always be a 1-1 ratio. After some heavy analysis, we discovered different relationships between batter and pitcher for the major event outcomes. With these pieces in mind, we then process the numbers using the Log 5 formula. In addition, if a batter "loses" homers in the matchup, these "lost" homers are distributed proportionally to doubles/triples.
Rest Hierarchy 1: When the 2nd in the hierarchy is to start, but gets scratched due to fatigue, will the 1st in line go in, or the 3rd, granted they're available? (ArlenWilliam - Hall of Famer - 3:31 PM)
When it's time to evaluate the hierarchies, the manager always grabs the players from the hierarchy in order from top to bottom. As long as they are eligible to enter the game (fatigue versus autorest), then they'll be selected. For player rest, there are no other criteria. For Def Reps, the player has to actually be a better defender. For PH, he has to be a better hitter.
could you please bring back diamonds legends baseballs past. i loved that format and it was easyier to draft and play. i am sure many people would sign up,including me. i loved that format. thanks bill renner (92336 - Hall of Famer - 3:36 PM)
When we were contacted regarding the Diamond Legends game, we compared the two games and decided what it would take for us to make a career based game. We didn't want to make drastic changes to our existing game, but still wanted to make a fun, career based game. That's what we did. There are obviously some big differences between the two. We generate a player's career statline so that his # of plate appearances and/or innings pitched dictate usage rather than injuries. We use a waiver wire system rather than a replenishment fund. We use the entire history of the game rather than breaking it down into eras. Aside from those differences, we offer much much more. Our theme league format offers an unbelievable number of league configurations. We're very excited and happy to have the former Diamond Legend users coming to WhatIfSports because we know they are a passionate, loyal, fun bunch. The biggest obstacle in making the change is dealing with change itself. It takes a season to learn the ropes, which is why we offered the free team. Our Players Guide located throughout the office provides a rather helpful look at how the game is played. Our forums offer lots of guidance and our support staff is always here to help.
I'd like to applaud your decision to remove the individual baserunning settings - any plans to further streamline and make the game less micro-managey? (bosoxbill - Hall of Famer - 3:37 PM)
We're always up for simplifying things as long as the simplification makes for a better game. If you and others have suggestions, please let us know via customer support.
Is there any way to get a 0 setting for the call bullpen? or another setting that we can use to ensure a starter pitches a CG no matter what? This is extraordinarily useful when trying to rest your bullpen, or other starters. (just4me - Hall of Famer - 3:38 PM)
We haven't done so yet, because you can accomplish this via other settings. For instance, if you want an SP to throw the entire game, give him no max pitch count and throw all your relievers to rest.
I'm gonna step out for a little while and let sbob85 tackle some questions. I'll be back in a little bit. ( Moderator - 3:39 PM)
Speaking of streamlining, what do you feel is added by the addition of "target" pitch counts? Feel free to lump this in with my last question. (bosoxbill - Hall of Famer - 3:45 PM)
Target Pitch Counts are meant to act as the Max Pitch Count of the previous version did. This number is the number that you would like him to get to. If he is doing very well or is about to close out the inning or other situations where it might be useful for him to stay in and he hasn't reached his Max Pitch Count, he can keep going. If he isn't in a situation where he will better the team and someone else could come in and be more effective, then he will be taken out. Just like in real life baseball, many pitchers are set to pitch counts that the manager wants to see them get to and pitch counts that they don't want them to exceed. These pitch counts are our Target and Max Pitch Counts.
There seems to be very little difference in "good" and "bad" plays between wide ranges of grades for most positions, especially in the middle infield. A B+ SS and a C- SS, for example, seem to produce almost the same split. Any chance that will be looked at by site staff? (JohnGPF - Hall of Famer - 3:47 PM)
Plus and minus plays are in part determined by a fielders values. A player with better grades should in theory have better plus and minus plays than a fielder with worse grades. This may not always be the case due to randomness and chances though. We will be willing to look into this though and make sure that everything is working as it should.
Rest Hierarchy 2: When the same player is in line to enter a game for two different positions, how does #$@# (pardon) Sparky decide where he goes? (ArlenWilliam - Hall of Famer - 3:49 PM)
Defensive replacements are done in the order of Bill James' defensive spectrum.
I just had a stretch where Mike Hampton threw 8 innings (2 runs, 90 PC), 9 innings (0 runs, 106 PC), & 9 innings (0 runs, 114 PC) in 3 consecutive games, each game starting at 100%. It's late in the season, and he is well below his RL IP, but I was under the impression that this sort of thing wouldn't be possible under the new system? Is the real change that you cannot "bank" IP for the playoffs, but you can still "bank" during the regular season? (zbrent716 - Hall of Famer - 3:53 PM)
The fatigue system for the regular season did not change with this update. The playoff fatigue system did change though. At the very beginning of the playoffs, everyone is set to having used up 95% of their plate appearances or pitches. If they have used up more than 95%, there is no update made. This stops one from storing up pitches for the playoffs to use low usage players from dominating in the playoffs.
Any chance of reversing the way Open and Champions leagues work -- such that users who have 5+ WS appearances are limited to only CLs, and those with less than 5 WS appearances have the choice between OL and CL? (just4me - Hall of Famer - 4:00 PM)
We are going to be keeping the Champions and Open leagues as they are. We want to allow users who have earned the right to be in Champions Leagues to either play in Open leagues and receive partial reward points or play in Champions leagues where they would have to prove themselves but would also get full reward points. The decision was made to allow this so that they didn't feel restricted to one type of league and feel inclined to possibly create aliases to play in Open Leagues if they really wanted to play in them. We also wanted a type of league that you had to work to get into and that it would be a true accomplishment to say that you were in a Champions League. Being eligible for a Champions League is a way of showing you are one of the top players. Competing in one and being successful would become much more of an accomplishment possibly then winning an Open League because in the Champions League, you know you are playing against the top players.
To follow up on my doubles and triples question because I'm not sure I understand the answer. Will, all other things being equal, a hitter with 8 doubles per 100 AB in a season when the league average was 5 be likely to hit less doubles than a hitter with 8 doubles per 100 AB in a season when the league average was 2? (JohnGPF - Hall of Famer - 4:03 PM)
Doubles and triples are normalized so what you are saying is correct. Someone who hit 8 doubles per 100 AB when the league average was 5 would hit less doubles (on average) then someone who hit 8 doubles per 100 AB when the league average was 2.
Re the base running agressiveness answer: Are you saying that if I set my team to very aggressive, sparky will know to only let my speedy base runners be very agrressive and will treat my 40 speed catcher the same as if the team level was set to rare? (JohnGPF - Hall of Famer - 4:06 PM)
Having a team base running aggressiveness of very aggressive will increase the number of chances of all of your players but the faster ones will have a significantly higher increase in chances than a very slow player.
Champions leagues take quite a while to fill and when they do it is populated with many of the usual suspects - this lack of variety leading to a less than enjoyable outcome for those involved. Any plans on changes for these leagues? (SpotSell - Hall of Famer - 4:07 PM)
This is similar to a previous question. There are no plans on changing the Champions leagues right now.
Is there any plan to bring the Defense options (CI; GL; OD; etc.) over to SLL? (zbrent716 - Hall of Famer - 4:11 PM)
Those items along with completely redoing LIVE are potentially in our pipeline of things to do.
You said "For player rest, there are no other criteria. For Def Reps, the player has to actually be a better defender." How is being a better defender determined? Straight fielding percentage or does range factor in? (SpotSell - Hall of Famer - 4:14 PM)
For most positions it is a weighted factor of fielding and range. For catchers it is a weighted factor of fielding and caught stealing.
It is probably too soon to be asking this, but are there any plans to introduce normalized fielding grades/rates at any point in time, such as the next update ;-)? (just4me - Hall of Famer - 4:19 PM)
Fielding percentage isn't normalized and won't likely be normalized. Range factors use our new relative range formula which is very similar to normalized. We like the way that the relative range factors turned out so we most likely won't be changing them anytime soon.
Does the OFer's defensive capabilities play any role in throwing out an advancing base runner, or is it strictly a matter of the runner's speed? (JohnGPF - Hall of Famer - 4:24 PM)
OFer's abilities do not currently play a role in throwing out an advancing base runner. This is something we would like to add in the future once we can collect enough data and analyze it properly. Right now, runners getting thrown it is based off many things including a threshold of historical numbers of runners thrown out and then many in game situational factors, i.e. speed, outfield in/deep, was the fielder running away so he had to turn and throw, etc.
Do any pitcher factors determine whether a batted ball will be a ground ball or fly ball? If so, what? (ArlenWilliam - Hall of Famer - 4:30 PM)
We were able to come up with GB/FB ratios for batters through data we could collect and analysis. For pitchers, this data was not available so a historical value is taken and used for all pitchers. There are also some situational factors that go into determining a GB/FB. An example would be that if there is a hit and run, there would be a higher chance of hitting a groundball.
How concerned are you that recent changes to the open leauge structure may not be achieving the desired effect, and providing a place for new users to play without being beat up by the hard-core "old guard?" Any plans to review? (bosoxbill - Hall of Famer - 4:36 PM)
I'm not sure if I follow that entirely, but we thought that adding a Champions Open League would help draw some of our more experienced users away from traditional Open Leagues. This would allow them to continue to receive full reward points/credits while also reducing the level of competition for new users in regular open leagues. As far as whether that's working or not, we'd say that it is to a certain extent but not as dramatically as we thought it would. It was something users have been asking for for quite a while, so we still believe it's a valuable addition to the preset league formats.
Do you realize your job is to give us a tool-set, for managing baseball games? (ArlenWilliam - Hall of Famer - 4:37 PM)
Why give us the ability to set a defensive hierarchy, when you override it by choosing the player you think should enter the game? (ArlenWilliam - Hall of Famer - 4:39 PM)
We want the defensive replacement hierarchy to represent a hierarchy of defenders, from best to worst. This is easy to verify on our end since it's based on how the stats are used in the engine. We weight them accordingly. By making sure that a defensive replacement from the hierarchy is actually better than the incumbent, all we're doing is fool proofing a system. Not putting this check in may be fun for some of the guys that run the theme leagues where the goal is to lose, but other than that it helps protect owners and gives them the best lineup in the field for defensive purposes.
How are secondary defensive ratings for positions at which a player has only played a few games determined? (JohnGPF - Hall of Famer - 4:42 PM)
Anybody who appeared in less than 20 games at a position has that position as a secondary position. From there, the player's primary position's defensive stats are mapped to the secondary position (which assumes his primary position's defense is more indicative of the type of defensive player he is than his limited time at his secondary position). We then whack his stats in a non-linear fashion based on the number of games under 20. A player who appeared in 19 games at 2B is impacted less than a guy who played 11 games at 2B. His stats then dictate his letter grades. Same thing for catchers who appeared in less than 20 games for their CS Pct.
Would you ever consider increasing the rewards - just slightly - for Champions Leagues? (bosoxbill - Hall of Famer - 4:43 PM)
Anything is possible. I guess it's something we'll review after the Champion Leagues are out for a while longer. Probably a year out or so.
Any speed runner on third in any game score/inning situation during a wild pitch; does the base runner (1) attempt to go home every time and (2) score every time? Ditto question for passed ball situation. What determines a steal home outcome; the catcherís arm rating? (team_acdc - Veteran - 4:46 PM)
Currently, we assume every wild pitch and passed ball in the game results in a base for those runners on base. We can always expand upon that.
Are y'all willing to entertain the idea that the Champions League concept might not be as good an idea as a less exclusive Veterans League concept? (JohnGPF - Hall of Famer - 4:47 PM)
Sure. The Champs league requires 5 previous World Series appearances. We believed this to be the right restriction. If it becomes apparent that it is too restrictive, we'll consider adjusting it.
What if I want to change the settings for particular baserunners on a per game basis, without changing the team style? (The same for removing for pinch hitter or pinch runner.) We're not paying "the manager" to control factors we want to control as players of your game. (ArlenWilliam - Hall of Famer - 4:52 PM)
What this came down to was complexity. When we add new areas to the game, we need to consider removing items that can be eliminated with better AI. If not, new users will immediately be turned off. With the latest update, we introduced a TON of new settings and stats. We reviewed every setting we had in the game. Based on previous answers in this chat, we believe the individual PH/PR/Baserunning settings can be simplified and accurately represented by a team-level setting. It's not taking away control, it's simplifying control into one location. The AI knows what you're trying to do because it has the same underlying stats that you're looking at. It knows that Carlos Beltran should try and take the extra base when Michael Barrett shouldn't. It knows that when you change your team baserunning setting from 3 to 4 that it wants Beltran to be more aggressive than Barrett. Beltran may attempt to take the extra base 10 more times and Barrett only once. We want to lessen the load you guys have to do to manage teams. Streamlining settings accomplishes that and doesn't give up any control. Like was mentioned earlier, if you run into boxscores where you disagree with a decision based on your team setting, let us know and we'll review it.
Is there any chance that in the future a form of the GIDP Opportunites stat will be added to the defensive stat pages for both teams and individual fielders? or is that just too complicated to manage? (JohnGPF - Hall of Famer - 4:53 PM)
That's something we can do. We track it, just don't display it.
What is the hierarchy and/or determining factor for the sim to bring in starters in relief? For example, since the update, only once has one of my starters come into a game he didn't start (although the relief box is checked) yet mop up and other pitchers have been used. So I guess I am wondering how the sim decides when a starter comes in; is it only after all other pitchers in the pen are gone? (alebenta - Hall of Famer - 4:54 PM)
Good question. We have an update going out today prior to the PM2 run that fixes an issue with this. If no reliever can enter the game due to settings and a SP can, then the SP should go in. This wasn't working 100% properly. It will be later today.
Why canít I see or determine in-game fatigue for pitchers at the point they left the game? Canít this aspect be shown in advanced game scorecard or somewhere? (team_acdc - Veteran - 4:55 PM)
We've never heard too much about wanting this in the boxscore. Adding it wouldn't be too difficult. As it stands now, you can't get a good feel for how in-game fatigue works until you play LIVE. It's what you'd expect , though. Once a pitcher reaches his limit (which is based on his pitches thrown for the season / games pitched), he begins to fatigue and becomes less and less effective.
"It's not taking away control, it's simplifying control into one location. The AI knows what you're trying to do because it has the same underlying stats that you're looking at." --- Comment, not question: WIDR, no, you don't know what I'm necessarily trying to do. Please let the auto-pilot guys have their defaults, but give us options to override per game/individual, thanks. (ArlenWilliam - Hall of Famer - 4:57 PM)
I'm just not seeing your point with these questions. If you could provide some examples of situations where what you want makes sense, we could better comment on them. Rather than clog the chat, please use the support system.
2 catcher questions. 1) I have yet to see a catcher make a "good" or "bad" play---is it impossible for them to do so? 2) does the SIM use a catcher's historical passed ball rate, and if not what is used to determine the likelihood of passed balls? (JohnGPF - Hall of Famer - 4:58 PM)
Currently, it is. We'll be improving this. We'll be posting some stats in the forum following the chat. We have catcher PB numbers, so it's based on their actual stats when applicable. When not available, we use a historical average.
Will you be evaluting the relative effects of your RRF numbers soon? It seems very high range outfielders still dominate the game, while infielders have very little impact. (ArlenWilliam - Hall of Famer - 4:59 PM)
Definitely. Be sure to check out the stats we post in the forums following the chat. It highlights some shortcomings we'll be addressing.
How are catcher CS rates determined for years when the stat wasn't kept? and how are we to determine the the relative ability for throwing out runners for catchers for whom no such stats (none pre-1960, i think) are included on the player pop up? Especially considering A+ arms go from under 40% to over 60% CS rates? (JohnGPF - Hall of Famer - 5:00 PM)
Our resident statisticians ran some statistical tests and derived forumulas by three different eras. These results then dictate the rate at which runners steal against them and the CS rate.
A top range OFer in CF is capable of performing at a plus 50 good/bad rate while a top range SS does not appear capable of performing at half that rate. I know IFer range is supposed to increase DPs, but is that good/bad dichotomy really how the system is intended to work? (JohnGPF - Hall of Famer - 5:01 PM)
The stats we'll be posting today highlight some of this. We'll be working on improving this.
Intentional Walks: How does the SIM determine when to do this (preventative offensive or proactive defensive); does the present batterís BA and SLG matter when compared to next batterís BA & SLG or is this just setting the plate for any double-play? (team_acdc - Veteran - 5:02 PM)
There are many variables involved. The managerial setting opens/narrows the window. Game situation, outs, score, inning, hitter, on-deck hitter all play a factor.
That's it for the day. Thanks for all the questions. We'll have another dev chat in a few weeks for those that couldn't participate today.