Steal Attribute: Too Low?

Collapse

Recommended Videos

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • Jolly Roger
    Prince of Plakata
    • Sep 2011
    • 871

    #1

    Steal Attribute: Too Low?

    In Franchise Mode, one of the stats that isn't quite right is Stolen Base Percentage. Teams tend to range from 75% down to 50%. The average should be just north of 75%, with more of a 60-80% range.

    The steal sliders don't seem to affect simulations, but tweaking the steal ratings does.

    For years now I've wondered why so many players have terribly low Steal ratings. There are only a small handful of players in the 80-99 range, but a huge amount of players in the 0-20 range. Stealing isn't that difficult of a skill. Just because a player is slow doesn't mean he'd have trouble reading the pitcher and getting a good jump. It's hard to know this since it doesn't happen, but, one could imagine the bottom half of the league in speed would steal at at least a 50% clip if they picked good pitchers and pitches to run on. They just don't steal much because that wouldn't be a good enough rate.

    But with all of these low Steal players, you have to think they collectively are at or below 50% steals, which is what might be bringing the overall steal % down.

    I've had success in testing with raising the steal ratings. I'm going to try a system where, regardless of speed, I match a player's expected SB% with his steal rating. So the absolute worst "stealers" should go from a 0 steal rating to maybe 45-50. Mediocre "stealers" will get a 50-60 rating, and players who are at least average will be in the 70s.

    I'll try this with a few teams and see what happens. If it works, it might be a good thing to pitch to the Devs for future years, something they could globally edit. Not sure if any of the roster people have played with Steal rating before, but it's fun to mess around with.
  • DarthRambo
    MVP
    • Mar 2008
    • 6630

    #2
    Re: Steal Attribute: Too Low?

    Keep us informed on your findings. This is definitely something I'd be interested in doing within my franchise. Which attributes are you editing? Steal success only?

    Sent from my SM-G996U using Operation Sports mobile app
    https://www.youtube.com/DarthRambo

    Comment

    • tessl
      All Star
      • Apr 2007
      • 5683

      #3
      Re: Steal Attribute: Too Low?

      I think they have it about right. Speed and steal ratings in the 80's = good base stealer.

      I use manage mode and if you pay attention there are trends the CPU has regarding when they pitch out, when they never pitch out. I'm in the process of compiling a list of 100 stolen base attempts and the % of time the cpu pitches out on each count.

      The steal attribute is how good a jump a runner gets. You need both to have a high rate of success. Good speed but poor steal will be a guy who gets caught stealing a lot.

      This is manage mode. I can't speak for modes the user controls the players because I don't use that.

      Comment

      • JTommy67
        Pro
        • Jul 2012
        • 598

        #4
        Re: Steal Attribute: Too Low?

        If OP is referring primarily to the sim engine, then I have to agree. For my RBU application I've spent quite a bit of time looking at this and concur that percentages are on the low side. The sim engine does not appear (or it didn't last time I checked) to factor in the speed rating in determining stolen base success rate. In-game engine is obviously another story.

        This is connected to a larger problem: they haven't messed with the sim engine mechanics to any significant degree in many years, and this is noticeable when viewing pitchers' complete games, frequency of saves, run production per game (which, per team, is about half a run lower than it should be), batting averages for high discipline, low contact hitters (or vice versa), among others.

        With respect to stealing, the 75% success rate observed in 2021 is actually the highest in history since we have complete records. In the early 2000s, the average was around 68%, and I believe this is around the time they probably began calibrating the sim engine (in the previous incarnations of the game). They've never fixed it, which is why we're seeing rates that are too low.

        Year SB CS
        2021 1677 545 .755
        2020 885 292 .752
        2019 2280 832 .733
        2018 2474 958 .721
        2017 2527 934 .730
        2016 2537 1001 .717
        2015 2505 1064 .702
        2014 2764 1035 .728
        2013 2693 1007 .728
        2012 3229 1136 .740
        2011 3279 1261 .722
        2010 2959 1129 .724
        2009 2970 1133 .724
        2008 2799 1035 .730
        2007 2918 1002 .744
        2006 2767 1110 .714
        2005 2565 1069 .706
        2004 2589 1100 .702
        2003 2573 1132 .694
        2002 2750 1282 .682
        2001 3103 1408 .688
        2000 2924 1323 .688
        1999 3421 1519 .693
        1998 3284 1505 .686
        1997 3308 1564 .679


        Bumping up steal ratings can solve this, but will also make it easier to steal in-game (which I believe is too difficult for the user anyway)...but at least that can be fixed with the slider.

        I'd be interested to see what others can find out looking at this further.

        Comment

        • Jolly Roger
          Prince of Plakata
          • Sep 2011
          • 871

          #5
          Re: Steal Attribute: Too Low?

          Originally posted by JTommy67
          If OP is referring primarily to the sim engine, then I have to agree. For my RBU application I've spent quite a bit of time looking at this and concur that percentages are on the low side. The sim engine does not appear (or it didn't last time I checked) to factor in the speed rating in determining stolen base success rate. In-game engine is obviously another story.

          This is connected to a larger problem: they haven't messed with the sim engine mechanics to any significant degree in many years, and this is noticeable when viewing pitchers' complete games, frequency of saves, run production per game (which, per team, is about half a run lower than it should be), batting averages for high discipline, low contact hitters (or vice versa), among others.

          With respect to stealing, the 75% success rate observed in 2021 is actually the highest in history since we have complete records. In the early 2000s, the average was around 68%, and I believe this is around the time they probably began calibrating the sim engine (in the previous incarnations of the game). They've never fixed it, which is why we're seeing rates that are too low.
          I'd love to hear more about what you've discovered. I spent most of yesterday coming up with my own stat to attribute conversion chart for a bunch of different categories. I really enjoy this sort of thing. Now whether I'm going to feel like manually applying this to the whole league, I don't know. Plus, the attributes change on their own in Franchise with progression and regression, although the manual edits would probably hold up for a season or three until needing to be calibrated again.

          I was thinking the exact same thing regarding the sim engine. As you say, a lot of the sim stats are consistent with what we may have seen 10-20 years ago.

          For example, in addition to there being more steals (and at a lower %) in the sim than in current day MLB, there are also about 1 to 1.5 less strikeouts per game, less home runs, and as you noted, about half a run less in sims.

          It would be amazing for us sim Franchise players if the devs could give us the ability to edit attributes globally. NBA2K allows this; I can't imagine it would be very difficult to implement. That way we could, for example, lower all HR/9 by 5, increase K/9 by 5, increase steal by 20, decrease base running aggressiveness by 10, etc. We'd be able to fix a lot of the sim stats ourselves.

          For the record, increasing steal rating and decreasing base running aggressiveness did allow me to achieve sim stats more in line with today's MLB. I edited the steal and baserunning agg on a few teams, ran a full season, and those teams were up around 80% with a realistic number of attempts. But again, the dilemma is, do I want to edit these attributes for all MLB players, plus all the other edits I feel would help?

          Comment

          • tessl
            All Star
            • Apr 2007
            • 5683

            #6
            Re: Steal Attribute: Too Low?

            Originally posted by JTommy67
            If OP is referring primarily to the sim engine, then I have to agree. For my RBU application I've spent quite a bit of time looking at this and concur that percentages are on the low side. The sim engine does not appear (or it didn't last time I checked) to factor in the speed rating in determining stolen base success rate. In-game engine is obviously another story.

            This is connected to a larger problem: they haven't messed with the sim engine mechanics to any significant degree in many years, and this is noticeable when viewing pitchers' complete games, frequency of saves, run production per game (which, per team, is about half a run lower than it should be), batting averages for high discipline, low contact hitters (or vice versa), among others.

            With respect to stealing, the 75% success rate observed in 2021 is actually the highest in history since we have complete records. In the early 2000s, the average was around 68%, and I believe this is around the time they probably began calibrating the sim engine (in the previous incarnations of the game). They've never fixed it, which is why we're seeing rates that are too low.

            Year SB CS
            2021 1677 545 .755
            2020 885 292 .752
            2019 2280 832 .733
            2018 2474 958 .721
            2017 2527 934 .730
            2016 2537 1001 .717
            2015 2505 1064 .702
            2014 2764 1035 .728
            2013 2693 1007 .728
            2012 3229 1136 .740
            2011 3279 1261 .722
            2010 2959 1129 .724
            2009 2970 1133 .724
            2008 2799 1035 .730
            2007 2918 1002 .744
            2006 2767 1110 .714
            2005 2565 1069 .706
            2004 2589 1100 .702
            2003 2573 1132 .694
            2002 2750 1282 .682
            2001 3103 1408 .688
            2000 2924 1323 .688
            1999 3421 1519 .693
            1998 3284 1505 .686
            1997 3308 1564 .679


            Bumping up steal ratings can solve this, but will also make it easier to steal in-game (which I believe is too difficult for the user anyway)...but at least that can be fixed with the slider.

            I'd be interested to see what others can find out looking at this further.
            I don't know about those who use the controller but steals are not difficult in manage mode if you use a player with the correct attributes. If you are trying to steal with guys who have 70's attributes in speed or steal then you probably aren't having success. My top guys have 90% success but I only steal with guys who are 80/80.

            Comment

            Working...