Swimming World's top 12 master swimmers!!!!!!!!!!!!!

I don't like using the Masters records -- you should base it on the actual world records. The Masters records are not "mature enough". I will give you a quick example - Laura Val set the 100 Free long course record in a 1:02 - about 9 seconds above the women's open world record. She also set the 100 Back record in 1:14.4 - that is about 15.5 seconds above the women's open world record. She is a fantastic swimmer - but her Free times are way stronger than her Backstroke times. By "actual world records" I assume you mean those of Phelps et al? The problem with real world records is that they don't cover a very wide range of ages, making it very hard to use them to model the effect of aging, especially when you extrapolate out to 80+ years. I think using masters records to rate masters swims is a pretty reasonable thing to do. Some events may be are harder hit by aging, or lack of training, than others (for me it has been 200 fly). The problem you note with Laura Val's records is mitigated to some extent by an averaging effect over all the age groups. Using masters WRs (and US records for SCY) gives a pretty reasonable fit, particularly if you use robust regression and take a few other precautions. An example of a typical fit is given here. The biggest problem is still probably extrapolating to the oldest ages (error is magnified). Also, the curves get lower with time, so a given swim "degrades" over time. (Although that is pretty much the case in real life too...what was once a great time can become more commonplace over the years.) I have tried it out for awhile and the system seems to work pretty well, it predicts outstanding times pretty much where you expect it. The current parameters are a little dated -- ie, pre-B70, etc.

I don't like using the Masters records -- you should base it on the actual world records. The Masters records are not "mature enough". I will give you a quick example - Laura Val set the 100 Free long course record in a 1:02 - about 9 seconds above the women's open world record. She also set the 100 Back record in 1:14.4 - that is about 15.5 seconds above the women's open world record. She is a fantastic swimmer - but her Free times are way stronger than her Backstroke times. By "actual world records" I assume you mean those of Phelps et al? The problem with real world records is that they don't cover a very wide range of ages, making it very hard to use them to model the effect of aging, especially when you extrapolate out to 80+ years. I think using masters records to rate masters swims is a pretty reasonable thing to do. Some events may be are harder hit by aging, or lack of training, than others (for me it has been 200 fly). The problem you note with Laura Val's records is mitigated to some extent by an averaging effect over all the age groups. Using masters WRs (and US records for SCY) gives a pretty reasonable fit, particularly if you use robust regression and take a few other precautions. An example of a typical fit is given here. The biggest problem is still probably extrapolating to the oldest ages (error is magnified). Also, the curves get lower with time, so a given swim "degrades" over time. (Although that is pretty much the case in real life too...what was once a great time can become more commonplace over the years.) I have tried it out for awhile and the system seems to work pretty well, it predicts outstanding times pretty much where you expect it. The current parameters are a little dated -- ie, pre-B70, etc.