Professors Who Analyzed 1.6 Million Putts

[iacas]

As Tiger showed us today... Low fairway % and low GIR meant Tiger spent all day today "scrambling" to make par

You do realize he hit more fairways and more greens than average this week, right? It's his putter that let him down. He said so himself, and he's usually pretty honest about his game after the fact.

Just FYI. Lucas Glover, I think, was leading the field in putting average for the week and was way up there in GIR.

[ApocG10]

obviously short game is important, but when your sitting 3 off the tee, your obviously in for problems, even if your short game is excellent. Long game is definately important, because you need it to get inside 100 effectively.

[DaveLeeNC]

I have some data relevant to this conversation. I use my own "home-brewed" stats system (MS Access based) and this doesn't represent all the stats that I generate. But I have listed some of the ones that fit in this conversation.

I took another spin at this just for grins. Another stat that I generate from the data that I collect is "up/downs with par putting". I look at all the shots that are not on the green but are within x yards of the pin. If the next shot is not on the green then it is a 0% up/down. If it is (for example) on the green and 6' from the pin I look at my putting table which says that I should average 1.6 strokes to hole out. That converts to a 40% up/down. If I am within 5' my table says 1.5 so that is a 50% up/down. Within 3' says 1.1 so that is 90% up/down, etc.

I normally use 35 yards as my "up/down limit" (all shots from off the green but 35 or less yards from the pin are calculated). I changed that to 65 yards and did the following. 1) I treated my "full shot game" measurement as completely characterized by "Error at par" 2) I treated my "short game measurement" as completely characterized by my "Net Putting" and "Up/down with Par Putting" parameter (with the 65 yard limit). The nice thing here is that these two numbers have no cross correlation so they can be directly combined (sum of squares). So my Full Shot Game correlation with "Scores against par" is the same as before (0.67) My Short Game correlation with "Scores against par" comes out to 0.65. So basically my data indicates that the short game (as I defined it) and the long game (as I defined it) are equally important. FWIW. dave ps. For those of you who are not experienced with correlation factors, here is an interesting exericize. Assume that we have 3 numbers A which is a random number between 1 and 100 B which is a random number between 1 and 100 (independent of A) C which is the sum of A and B The correlation between A and C will be a tad over 0.7 The correlation between B and C will be a tad over 0.7 The correlation between A and B will be 0 In this particular case the correlation between A/C (squared) plus the correlation between B/C (squared) will be 1.0