DG Statistics?

[Cruz Duck]

Any interest in talking about DG statistics?

I was reading about some statistics used by ball golf and read that one of the strongest indicators of overall performance is "Greens in Regulation". This means if you get the ball on the green with at least 2 fewer strokes than par.

So I was thinking "Circles in Regulation" might be a good statistic for disc golf. How often do you get into the circle with one fewer stroke than par. Other statistics that might be interesting:

1) Birdie Opportunities - how often do you get birdie opportunities? (Within 50 feet of the basket with two strokes fewer than par?

2) Drives in Fairway - what percentage of your drives land in the fairway?

3) Par Saves - Your drive lands out of the fairway or you just have a poor drive. A great approach shot and/or putt saves a par. I could use a better definition here, but you get the idea.

4) I think you could create a single number that is shows how successful you were at putting during a round. Something like start with 100, add a few points if you hit a putt between 20 and 30 feet. Add more points if you hit a putt longer than 30 feet. Subtract points if you 3 putt. Subtract points if you 2 putt from 20 feet or less. I am not sure of the exact algorithm, but I think an interesting number could be created.

I have been unhappy with the android/iphone scorekeeping apps on the market, so I am writing one for my own use. I want to include some statistics in the app and I am considering the above stats. What do you think?

 

[Widdershins]

The stats I would be most interested in are related to putts, as putting tends to be the biggest variable in my game.

1) Total Putts Inside the Circle. We know that in a round of 18 that number should be around 18. On some holes there may be no putts (aces, upshots or long putts which go in) but it gives a rough idea of putting prowess. It is also objective. Every putt is either inside or outsdie the circle (so mostly objective as we have to guess whether some putts are inside or outside when the circle is not marked).

2) Plus or Minus Putting. This is the stat I sort of keep in my head to judge how well I putted during a round and roughly based on my personal putting comfort range. Any putt I should make is Zero (0) if I make it and minus 1 (-1) if I miss it. Any putt I don't expect to make (a bonus putt) is Plus 1 (+1) if it goes in but Zero (not counted) if it misses. For this stat, any round score at zero or better is great.

This Stat is subjective, based on my own expectations. In certain conditions (winds, obstacles) a putt even inside the circle is something I know I will not usually make, a bonus putt so to speak.

This stat is the most useful to me because I know my own skills. If I miss a putt I know I should have made it can be made up with a bonus putt (one I know is outside my comfort zone). Even if my comfort zone is far different than someone else's, it doesn't really matter as this is a personal stat, not meant to be compared with other players. The nice thing about this stat is that it is easy to keep track of. The number of missed easy putts and made bonus putts should both be pretty small making the math simple.

Driving stats like Fairway Percentage and Greens in Regulation are somewhat deceptive. A drive can hit an early tree and bounce in the fairway and still be a poor drive. A drive can reach the putting circle and still have no putting line on some holes. Some drives can land out of the fairway and still be excellent drives.

I tend to view holes in a Plus or Minus way. An easy hole which should be deuced counts as Zero if I deuce it and Minus 1 if I par it. A tough par hole is Zero for a par, Plus 1 for a birdie and Minus 1 for a bogey. Some holes play as actual par 4s or 5s no matter what they are called so I adjust the scoring accordingly. This method compares my results to the field I am playing against. In the Pro divisions if you par an easy hole you have lost a stroke to the field. So there are times you may string a bunch of deuces together but all you are doing is keeping pace with the field (well, the top of the field).

 

[Itchy1]

Mr. Ellis, your self-evaluation reminds me a lot of how the NHL keeps stats and grades their hockey players. I've actually been doing the +/- on my putting too but my points are little different than yours, I just keep track inside the circle.

 

[JHern]

There's a lot to like about Mark Ellis' system, but it would be nice to do something more objective, which would allow you to compare with others, and also give you a way to measure your progress over time.

Just for the sake of discussion, consider the complete data set for a round (if you were to record all of it), which would consist of a list of every throw you made, and its distance from the pin. (Define zero distance=in the basket/holed-out.) The tee shot entry will be the same regardless of what happens, so it's a given. If you get a deuce, you'll only need to actually record the distance for a single entry in the list. The quality of a shot can be judged by looking at the subsequent entry or entries...if you holed out on a throw, then there is no next entry (that hole was finished). If you had a long next shot, with little change in distance from the previous shot, then your previous shot was either poor or you had to pitch out of the shule from the shot before that (either way, the badness shows up in the list).

If you were to record all this, then you can judge the quality of any shot by taking the distance entry, subtracting the distance listed in the next shot (or use zero if you holed out), and then dividing again by the entry itself...multiply by 100, and you have the percent ground covered between you and the basket (this is 100% when you put it in the basket). I think this is a useful measure...I'm going to call it the "Q" of the shot.

If you were to make a plot of Q vs. distance for all your shots (sorting them from closest to furthest), then you would be able to see how your ability to cover ground tails off as distance increases. At closer ranges, this would represent putting. You would see Q=100% at close range, and tailing off to Q<100% at further ranges. Inside your "magic circle" everything holes out...you could obtain the magic circle of your round from the list, as being the largest distance below which everything is Q=100%.

Depending on the number of holes, you can count the fraction of made putts in various distance ranges, say 0-10', 10'-20', 20'-30', 30'-40', 40'-50', and so on until you no longer have Q=100%. Then make a list. So, say you made 8/8 putts at 0-10', 9/10 putts at 10'-20', 4/7 putts 20'-30', 1/12 at 30'-40', 0/15 at 40'-50', 1/6 at 50'-60', and none beyond 60'. So you would report 100-90-57-8-0-17.

Anyways, I could go on and on, and this would be about analyzing all the data. But we're instead interested in trying to get something useful and objective without writing all this down and doing a sophisticated analysis...

...will be thinking about this, Steve. Fun stuff.

 

[Cruz Duck]

Hmm, I guess that it all depends on what you want from the statistics:

1) To compare your current round against yourself or against your competition. Did I do as well as I think I could? Or how did I do against what I think the competition will be doing? Mark suggested several statistics of this type. Thanks Mark.

2) To compare your performance against others.

3) To isolate specific skills. My score was good, but how was my driving? My putting? My approach?

4) To get an idea on how my specific skills relate to others. Or to compare/contrast two other players.

John seems more interested in the last three reasons. Thanks for the thoughts, John.

Maybe I should talk about what my motives are.

When I was a kid, I used to love to go over baseball statistics. It gave me a view of the major league players and how they were doing that year. For example, you get a different view of a center fielder who has a .305 batting average, 35 steals, 3 home runs and 35 rbis versus a right fielder who has a .275 batting average, 5 steals, 35 home runs and 105 rbis. Plus you could compare the statistics of the current year against career statistics to get an idea of how they are doing this year.

I think it would be great to have similar statistics on disc golfers. I know who most of the top golfers are by name and face. But I really don't know much about their skills. It would also be interesting to know if Nikko is on fire due to his putting or if Nate is going through a slump due to his drive accuracy.

Furthermore, I would like to be able to compare my skills against the pros and other ams.

Having said all this, I realize how hard this would be to do. At a baseball game, one guy sitting in the stands could keep the stats for all of the players. At a disc golf tournament, you would probably have to have a person at every hole keeping score or maybe a different person following every group.

Mark, I agree that drving stats like Fairways in Regulation or Circles in Regulation can be deceptive. But I think that is true with any statistic. It is a lot easier to hit home runs in Colorado than San Diego. It is a lot easier to get RBI's if you come up with men on base, etc.

Still I would like to work on some stats and gather data, even if it is only for my use. So I plan to create an Android and/or iPhone app and gather data from my rounds and my friends rounds. I would also like to follow the lead group at the Master's cup and may some other local tournaments to see what kind of numbers come from the top pros at the same courses that I play.

The only way to get this data is with a smartphone app. I can't see people writing the data down and going home and entering it into a computer. I don't expect to see people lugging a tablet around the course. It has to be quick and easy to enter the data on a single smartphone screen. I am leaning towards things like Drives in Fairway, Circles in Regulation and Birdie Opportunities, because each can be implemented with a single checkbox. I also like Mark's idea of Total Putts within Circle, so I will try to figure out a way to get it in.

If this looks promising, I might release the apps for free and set up a server in the cloud. Then players could upload their scores/stats to the server and then anyone could look at the numbers.

Yikes, this sounds like a lot of work, but I want to get started on it and see where it goes.

Any other ideas, thoughts, comments are welcome

 

[ChrisWoj]

Mr. Ellis, your self-evaluation reminds me a lot of how the NHL keeps stats and grades their hockey players. I've actually been doing the +/- on my putting too but my points are little different than yours, I just keep track inside the circle.

This is my system, pretty much. For every miss inside of the circle it is worth -1. For every make outside of the circle it is worth +1. I've noticed a direct correlation between my +/- and my 1000 rated rounds. If I am in the positive - I shoot over 1000, if I'm dead 0 I am right on 1000 rated, and if I'm below 0 I'm below 1000 rated for the round.

 

[JR]

Going spreadsheet deep in cataloging everything is gonna take away from the most important stat, score, if you do all the possible stats on the go, while playing. And it makes you a slower thrower.

It is easiest to go deep with the math from a video footage afterward. There are few vids, that show each throw of a group for a round and fewer still for an event. I think each shot Feldy made was published in the SDGO 2010 vid and i think every shot in the final by every lead card player made it to the published version. Not all shots are shown on lcgm8 videos on Youtube.

I'd like to hear about your perspective on being filmed for a round, because we get requests from players to not show some shots so often that we can't remember them all. So occasionally bad shots are on the screen. It took a player the caliber of Feldy to not have terrible errors on screen being filmed shot for shot. Are you up to that level and would you take extra pressure from knowing your every move is on Youtube forever? Oops

 

[Steady 26542]

In regards to driving, what if you put down L, R, or C. Meaning, you threw your drive (L)eft of where you wanted it to go, (R)ight of where you wanted it to go, or (C)orrectly. As Mark Ellis said, your drive might have hit a tree to the right and then came back to the fairway. By logging where you threw your drive you get a pattern of where you are throwing.

 

[harr0140]

I have kept my statistics from day 1 (on another DG website) and whenever I make a shot outside of the circle I write the distance down. I also always keep track of any putt I miss inside the circle so that goes down as a 2 putt. This would also be able to tell me if I have "hit the green" in regulation. Any 2 putt par was on the green, any 1 putt birdie was on the green. Anything with a zero putt was not on the green

My first year I averaged 1.18 putts per hole (about 21 per 18 hole round) and I am now down to 1.14 after 3 plus years of playing.

 

[Cruz Duck]

I used to keep some type of record of every shot. For example, I might write:

1 – x – 45 – 15

This would mean hole 1, excellent drive, missed 45 foot putt, made 15 foot putt. Or

13 – p – vg – vg – 25

This would mean hole 13, poor drive, two very good approach shots and made a 25 foot putt. My plan was to put this data into a spreadsheet and then run some statistics. My hope was that over time I would see the percent of excellent drives go up. Instead, I ended up with a bunch of sheets of paper with scribblings all over them. I now hope that I can capture data like this on a mobile app and it will create the statistics for me.

I have completed the first version of an Android app and tested it on a tablet. I was getting ready to go buy an Android phone, when I got an old mac laptop. I have started porting the program for an iPhone. Unfortunately, I got diverted with another project, but I am about ready to start working on the dg scorekeeper again.

I did find some interesting things while testing the Android app. I created a driving/approach statistic and a putting statistic. I expected the driving/approach statistic would vary by the course. It is much easier to keep your drives in the fairway and get them into the circle on an easy course. On the other hand, I expected the putting statistic to be pretty constant across courses. A 25 foot putt is just as hard or easy on any course.

However, I found that the putting statistic varied as much as the driving/approach statistic. I found that on hard courses I often would have a good drive, then 50 to 75 foot approach shot which left a 10-15 foot putt. I made almost all of those putts. On an easy course, I would make a good drive and be left with a 25 to 30 foot birdie putt and my putting scores dropped. Therefore, I had better putting scores on hard courses than easy courses.

I should point out that I am an Am 2 player. When I have played with very good players, their experiences are different. On easy courses, they are driving within 20 feet of the basket. On hard courses, they are left with 25 – 40 foot birdie putts. I have not scored any of their rounds, however I would expect them to have better putting scores on the easier courses.

 

[Apathy]

Dgcr scorecard is awesome for the putt stat. Anything outside the circle I put down as zero putts, the goal being to keep your average putts a hole at 1 or under. So far I am at 1.06

 

[JHern]

cruz duck, have you ever talked to Steve Ganz? I think he's pioneering a similar effort for PDGA. You guys should definitely talk, if you haven't already.

A round of golf (ball or disc) is typically non-linear in score. That is, if you're playing well, then you'll get closer to the pin, and you'll make longer putts. Your score can swing dramatically from bad to good rounds, because good throws build on good throws in good rounds, while bad throws compound bad throws in bad rounds. The more holes you play, the better your average score represents your true skill at that time. That's why it's harder for anyone to shoot 1100-rated on 27 vs 18 holes (18 hole correlated fluctuations are far more common than 27 hole correlated fluctuations). But what I think is also very interesting is the dispersion and median. These can tell you something about your game, and maybe help you to figure out a way to improve it.