Fair assessment but look at this way, you seem to like a heavy racket of 376 g and a balance of 32.8. Any racket you hold steady in your hand that weighs 376 g and has a balance of 32.8 cm will feel the same. You will only notice a difference in when you swing the racket. If you add mass at 7" the racket will get heavier the more weight you add and will feel heavier in your hand. Also the SW will go up with every gram you add. But with if you your original intent was to modify your racket so it weighed 376 g, and a balance of 32.8 cm, and an Mgr/I of 21? Well for one thing the racket will swing faster because the only way for you to get an MgR/I of 21 with your weight and balance specs is to have a SW of 366 which is 20 points lower than what you have now. So if your original intent was to modify your racket to 376 g, 32.8 cm, and SW of 366 you would never have to add weight at 7" to get the MgR/I you want.
True but if you consider risk and reward I think you'll find Federer thinks the reward was worth all the problems.
Yes but modifying a racket is not a tedious process and if you have sound knowledge of what you're trying to get it's actually pretty easy.
My guess is P1 decides, "Hey, today is a nice day. Lets get 9000+ racquets with different specs and see which one Andy likes best!!!"
Sorry, meant few hours perday for a couple of weeks.
So does layup change feel?
You should post all results because statistical tests can produce false positives by chance. If you performed many tests and got many insignificant results, your significant results may not be valid.
I have posted standard errors (approximately 90% confidence intervals). This way, you can decide for yourself whether you deem the results significant or not.
Your results are significant by the definition of statistical significance. But without context statistical significance doesn't mean anything. For instance, I can find statistically significant results in random data if I write a few lines of code to search for them. I'm not saying your results aren't valid, but unless you disclose all your work, there's no way of knowing.
Travlerajm, I want to thank you for all this work and research. I have said before and will say it again, I don't believe in mgr/i but you do, and the analysis you bring are always interesting and I always enjoy reading them.Chris was a playtester for 67 of the racquets that currently appear on the Tennis Warehouse site with individual playtest scores.
I analyzed his component rating scores and observed the correlations between his preferences and various racquet parameters. For Chris, Control, Touch, and Volley scores showed an identical trend with respect to MgR/I value, with a distinct optimum value at about 21.0. The combined average trend is shown here - error bars represent standard error. The second chart shows the trend for topspin, which - for Chris - is essentially the reverse of the trends for control:
For background, a racquet's MgR/I gives a relative measure of a racquet's angular acceleration about the wrist access, assuming a double pendulum swing model. Tuning the MgR/I value to match the timing of player's natural swing (so that the player can relax the wrist through the hitting zone without needing to apply any torque from the wrist muscles) gives enhanced control.
M = mass in kg.
g = 980.5 cm/s^2
R = distance from butt end to balance point in cm
I = Swingweight moment of inertia about butt end of racquet.
For a related thread:
https://tt.tennis-warehouse.com/ind...-review-ratings-vs-mgr-i-measurements.559553/
For groundstrokes and returns, Chris was very sensitive to a racquet's MR^2 value. Every single racquet that Chris rated highly was clustered in a narrow apparent optimal range with MR^2 value near the low 350s:
You might check out the link provided in the original post of this thread. There I compiled mgr/I data nearly 200 racquets and plotted the average tw playtester score for all of the play test components.
There is a big difference in performance between a frame with optimized mgr/I and one that has mgr/I value merely 0.1 away from the optimum. Assuming that 21.0 is the optimum for a given player, then you would need to add a whole lot of lead tape to the handle to convert a typical (20.7) stock racquet into an 'optimized' one.
Maybe that's the process @movdga was talking about when he said it was just too complicated. Assume you have a racket you like and the balance on 32 cm but you want to raise the SW up from 301 to 320. That's pretty easy add 5.85 grams to 12 o'clock and you're up to 320. But that weight raised your balance point up but it's easy enough to add the mass you need in the butt to drop the balance down to 32 cm. But if you have 2 or more racket and the specs start out close but not the same getting them all to the same mass, balance and SW becomes a little more involved but not too difficult. There are apps that will tell you what mass to put where to do that. Then you want to jack up the MgR/I from 20.7 to 21.00. What's going to happen to the overall balance? It's going down and to get it back you must increase the SW which throws off your MgR/I all over again. Yeah I think the process could be easier. If you look at the formula there must be some mass (m) that you can shoot for that will give you a balance of 32 cm, SW of 320, and an MgR/I of 21. I wonder what it could be? (HINT 335.40 g) Let's see what happens if I shoot for 335.4 g, 32 cm, and SW of 320
Or you could simply use a spreadsheet.
Did you choose Chris because he was the only playtester with available data, or did you choose him because you couldn't get significant results with other testers? Also breaking continuous data (in this case the MGR/Is) into categories isn't good statistical practice. A better way is to decide what an optimal MGR/I is ahead of time and then compute for each data point the magnitude of the difference between the actual and optimal MGR/Is. Then regress that magnitude on the review score and check the coefficient for significance. Finally, you need to control for other variables in your regression (e.g. weight, balance, swingweight, etc.), or you can't tell whether it's another variable correlated with MGR/I that's actually affecting the reviews. As it stands now, your results are intriguing, but until you do a more robust study, you can't claim anything about MGR/I affecting performance.
I did
I chose Chris simply because he has reviewed the most racquets.
Mass = 272.16 g = .27216 kg
A polarized frame is a frame that has more mass concentrated to the tip and But than in the center. To some degree all tennis frames are polarized and the more weight you add to the ends the more polar it becomes. The more mass you add to the balance point the less polar it becomes. A polarized frame is a frame with higher Recoil Wieght. The higher the RW the more polar it is.
I is 11
