Alex Table Tennis - MyTableTennis.NET Homepage
  Help Desk Help Desk  New Posts New Posts RSS Feed - Handicap formula for 11 points
  FAQ FAQ  Forum Search   Events   Register Register  Login Login

Handicap formula for 11 points

 Post Reply Post Reply
Author
jrscatman View Drop Down
Premier Member
Premier Member
Avatar

Joined: 10/19/2008
Status: Offline
Points: 4585
Post Options Post Options   Thanks (0) Thanks(0)   Quote jrscatman Quote  Post ReplyReply Direct Link To This Post Topic: Handicap formula for 11 points
    Posted: 04/18/2012 at 12:18pm
Hello,

Are there any handicap formula for 11 games. Our club is planning on having a handicap tournament. But the person who runs it just assigns 3 point, 4 points or whatever to people. Clearly, doesn't work - because the top players always end up winning the tournament.

Was wondering if there was a standard way of assigning handicaps for players - based on ratings.

Thanks in advance
JRS
Butterfly MPS
FH: Donic Acuda S1
BH: Palio CK531A OX
Back to Top
Sponsored Links


Back to Top
ttran82 View Drop Down
Super Member
Super Member


Joined: 01/21/2011
Location: US, Bay Area
Status: Offline
Points: 133
Post Options Post Options   Thanks (0) Thanks(0)   Quote ttran82 Quote  Post ReplyReply Direct Link To This Post Posted: 04/18/2012 at 3:08pm
A handicap system I saw at one of the club was 1 point for every 100 rating points.
Hope this helps
Back to Top
vvk1 View Drop Down
Gold Member
Gold Member
Avatar

Joined: 11/14/2009
Location: UK
Status: Offline
Points: 1883
Post Options Post Options   Thanks (0) Thanks(0)   Quote vvk1 Quote  Post ReplyReply Direct Link To This Post Posted: 04/18/2012 at 3:27pm
In our club, all players are assigned a rating before the handicap tournament. 

The rating  is based on player's division and win percentage during the season. The formula looks like this: base_ranking + win_percentage * 2. Base ranking is increased from lowest division to highest. So 4 division uses 50 + wp*2, 3rd division uses 100 + wp *2, 2nd division uses 150 + wp *2, 1st division uses 200 + wp *2, prem division uses 300 + wp *2. Note that this allows for overlap between divisions, so a 80% player from 2nd division will be about the same as a 55% player from 1st division, and so on.

Then there is a table which allocates handicap points based on rating difference. E.g.
Rating difference      Handicap Points
0-3                            0
4-8                            1
9-15                          2
..
117~142                  10
...
461-500                   18


Back to Top
jrscatman View Drop Down
Premier Member
Premier Member
Avatar

Joined: 10/19/2008
Status: Offline
Points: 4585
Post Options Post Options   Thanks (0) Thanks(0)   Quote jrscatman Quote  Post ReplyReply Direct Link To This Post Posted: 04/20/2012 at 2:31am
Thanks guys. 

Are your handicap tournament games - 21 points or 11 points or some other game score. 
Butterfly MPS
FH: Donic Acuda S1
BH: Palio CK531A OX
Back to Top
Thomasson View Drop Down
Gold Member
Gold Member
Avatar

Joined: 03/30/2010
Location: Netherlands
Status: Offline
Points: 1005
Post Options Post Options   Thanks (0) Thanks(0)   Quote Thomasson Quote  Post ReplyReply Direct Link To This Post Posted: 04/20/2012 at 2:39am
We have this handicap tournament in between seasons, this is the handicaps we use:

1     2     3     4     5     6     7     8     9    10 verschil in aantal klassen

2     4     5     6     6     7     7     8     8     9  verschil in aantal punten

First one is different in classes, in the Netherlands its like First class/second class etc.

Second row are the differences in points between each class.

So if these are your players.

... = level a

... = level c

... = level d


A vs C starts with a score of 0-4 each set.
A vs D starts with a score of 0-5 each set.
C vs D starts with a score of 0-2 each set.


You just count how many classes are in between level of player A vs player B and then check the schedule.


Maybe you could use this with ratings or classes yourself, they changed it this season and I think it will be perfect.





Edited by Thomasson - 04/20/2012 at 2:42am
Blade: TB ALC
FH: Tenergy05 2.1
BH: Tenergy64 1.7
Back to Top
vvk1 View Drop Down
Gold Member
Gold Member
Avatar

Joined: 11/14/2009
Location: UK
Status: Offline
Points: 1883
Post Options Post Options   Thanks (0) Thanks(0)   Quote vvk1 Quote  Post ReplyReply Direct Link To This Post Posted: 04/20/2012 at 3:15am
Originally posted by jrscatman jrscatman wrote:

Thanks guys. 

Are your handicap tournament games - 21 points or 11 points or some other game score. 

21 points, best of 3.
Back to Top
smackman View Drop Down
Assistant Moderator
Assistant Moderator
Avatar

Joined: 07/20/2009
Location: New Zealand
Status: Offline
Points: 3242
Post Options Post Options   Thanks (0) Thanks(0)   Quote smackman Quote  Post ReplyReply Direct Link To This Post Posted: 04/20/2012 at 11:02am
we used to have them up to 51 and novice starting on 46 and superstars even at -10
and still got many close games

 I would suggest having a longer game up to say 21 and then fitting in the best and worse players at each ends of 0  to 20 and then big fat guesses
also random draws are important (for example the 2 best players may be drawn to play each other in the first round, i.e no seedings
Ulmo Duality,Donic BlueGrip C2 red max ,Yinhe Super Kim Ox Black
NZ table tennis selector, third in the World (plate Doubles)I'm Listed on the ITTF website
Back to Top
jrscatman View Drop Down
Premier Member
Premier Member
Avatar

Joined: 10/19/2008
Status: Offline
Points: 4585
Post Options Post Options   Thanks (0) Thanks(0)   Quote jrscatman Quote  Post ReplyReply Direct Link To This Post Posted: 04/20/2012 at 12:25pm
Thanks - great suggestions
21 -  2 out 3 with Random draws - 100 rating points = 1 pt. This seems simple and do able!
Butterfly MPS
FH: Donic Acuda S1
BH: Palio CK531A OX
Back to Top
tt4me View Drop Down
Gold Member
Gold Member
Avatar

Joined: 01/17/2013
Location: RC Poverty Zone
Status: Offline
Points: 1019
Post Options Post Options   Thanks (0) Thanks(0)   Quote tt4me Quote  Post ReplyReply Direct Link To This Post Posted: 10/04/2013 at 2:27pm
In a recent post Jay mentioned handicaps.  Instead of sidetracking that thread I am posting on this thread because it is on topic.

I once saw a table of handicaps for hard bat play.  Obviously it assumed the games were to 21.  I decided to make a chart for games to 11 and games to 21 so here it is
The first part shows the method I used.
The charts are on page 4 and 6.

The math or numbers should be perfect.  I didn't round anything like what I have seen on other tables. What can be debated is how the probabilities equate to ratings differences.  I used the USATT ratings chart. I used the fact that if a player A is to maintain a 100 points difference over player B then player A must win 5 games to 1 or 5/6 of his matches over player B.   I used this as a yard stick for the scaling of probabilities to ratings differences.

The tricky part was figuring out how to handle the case when a game gets to deuce.  You would think that to win a games it one must win 11 out of 20 balls or 55% but when deuce situations are taken into account one needs to win a fewer percentage.

I would trust these tables over others I have seen on the internet.  They appear to be arbitrarily rounded to look good.   I would be interested in what David Marcus has to say about this.

I play handicap games with my coach.  The purpose is to get to tight games to learn how to win the tight ones.  I can also estimate my rating relative to my coach by using this table.


    
Back to Top
Baal View Drop Down
Forum Moderator
Forum Moderator


Joined: 01/21/2010
Location: unknown
Status: Offline
Points: 14213
Post Options Post Options   Thanks (0) Thanks(0)   Quote Baal Quote  Post ReplyReply Direct Link To This Post Posted: 10/04/2013 at 9:59pm
I think tt4me's post on this should be "stickied" somewhere for future reference, and maybe sent to USATT.  This is the kind of analysis that is really needed for these kinds of things.
Back to Top
benfb View Drop Down
Platinum Member
Platinum Member


Joined: 10/10/2008
Location: United States
Status: Offline
Points: 2709
Post Options Post Options   Thanks (0) Thanks(0)   Quote benfb Quote  Post ReplyReply Direct Link To This Post Posted: 10/04/2013 at 10:32pm
Originally posted by jrscatman jrscatman wrote:

Thanks guys. 

Are your handicap tournament games - 21 points or 11 points or some other game score. 

The USATT web site has schedules for handicap point systems. As I recall, there are systems for 32 points, 42, points, and 51 points (give or take a bit).  They are all intended to be played with a single game.

Some years ago our club ran a weekly league with handicaps on 11 point games.  The problem is that 11 points is so short that it only takes a few edge/net shots to cause an upset.  I was one of the top players in that league and it was very stressful to me and took a lot of the fun out of it -- there was no room for bad luck.

More recently, our club organized a couple of handicap tournaments with a single game to 32 points.  My experience from that was that sometimes that just wasn't a long enough stretch.  I'm thinking of organizing another one and I'll probably use the 51 point scheme.

The thing to realize is that handicaps aren't always accurate.  Surely you've had cases where someone was rated far below you but still gave you a rough time? (And the opposite, where you do well against someone above you, is always pleasant.)  You need enough points to make up for the handicap.  The longer the same, the more chance for a fair result.
Back to Top
tt4me View Drop Down
Gold Member
Gold Member
Avatar

Joined: 01/17/2013
Location: RC Poverty Zone
Status: Offline
Points: 1019
Post Options Post Options   Thanks (0) Thanks(0)   Quote tt4me Quote  Post ReplyReply Direct Link To This Post Posted: 10/04/2013 at 10:54pm
Originally posted by benfb benfb wrote:

Originally posted by jrscatman jrscatman wrote:

Thanks guys. 

Are your handicap tournament games - 21 points or 11 points or some other game score. 

The USATT web site has schedules for handicap point systems. As I recall, there are systems for 32 points, 42, points, and 51 points (give or take a bit).  They are all intended to be played with a single game.

Some years ago our club ran a weekly league with handicaps on 11 point games.  The problem is that 11 points is so short that it only takes a few edge/net shots to cause an upset.  I was one of the top players in that league and it was very stressful to me and took a lot of the fun out of it -- there was no room for bad luck.
But you can see I figure the handicap for an even match, not an even game.  Sure one can get a few bad breaks and lose one game but that doesn't lose a match.

Quote
More recently, our club organized a couple of handicap tournaments with a single game to 32 points.  My experience from that was that sometimes that just wasn't a long enough stretch.  I'm thinking of organizing another one and I'll probably use the 51 point scheme.
I can change a few numbers and recompute for any length.  You can see that at the end of the pdf I toyed with 5 point games.  If you use 51 point games the spread in ratings points per handicap ball will be smaller.

Quote
The thing to realize is that handicaps aren't always accurate.  Surely you've had cases where someone was rated far below you but still gave you a rough time?
That is a problem with the rating system not the probability tables.

Quote
 (And the opposite, where you do well against someone above you, is always pleasant.)  You need enough points to make up for the handicap.  The longer the same, the more chance for a fair result.
That is why I did the tables for 3 out of 5 games.   The winner must win at least 33 points.   I can easily make a new table for  best of seven but that would make the matches longer.  I didn't think it would be necessary because any match that used handicaps wouldn't affect anybody's rating.

One more thing.  The ratings spread per ball is quite coarse when playing 11 point games.   It may be possible to 'refine' my table more so that to simulate a handicap of 5.5 balls the lower player gets a 6 ball handicap one game and 5 the next.   




Edited by tt4me - 10/04/2013 at 11:00pm
Back to Top
stiltt View Drop Down
Assistant Admin
Assistant Admin
Avatar

Joined: 07/15/2007
Location: USA
Status: Offline
Points: 816
Post Options Post Options   Thanks (0) Thanks(0)   Quote stiltt Quote  Post ReplyReply Direct Link To This Post Posted: 10/04/2013 at 11:00pm
Wow tt4me rocks!

The following is an empirical approach and I bet it can go hand in hand with the science he linked to.

It is a handicap system that may please 2 practice partners presenting a substantial difference in level and wanting to play meaningful games to recreate the pressure in matches.

We assume I am the weakest player and I have no chance to win a regular game against my practice partner:

We start playing a regular game and I start the next game with a lead corresponding to the difference in points from the score in the 1st game; let's say I lost 11-8; we start the next game 3-0 in my favor. That's only done once. In the 3rd game if I loose again I get one more point and start the 4th game at 4-0 in my favor; if I win we take one point off my handicap and I start at 2-0 and so on so forth.

At the "end of the day" after enough games we know what's my true handicap.

The system allows the 2 players to win close to 50% of the games; the weakest player will learn to close the deal under pressure and resist the comeback; the better one will push him/herself to come back. It is interesting for both for different reasons.

Maybe in a club with enough pairs of people willing to take notes over 20 games, somebody from all the reports could find a clear correlation with the USATT rating system to deduct from 2 ratings the corresponding handicap; then a comparison with tt4me tables will be fun.

Credit to Tom Veatch who taught me the system in his basement when I was progressing. That was fun! Big smile

Back to Top
ybok View Drop Down
Super Member
Super Member
Avatar

Joined: 10/06/2004
Status: Offline
Points: 104
Post Options Post Options   Thanks (0) Thanks(0)   Quote ybok Quote  Post ReplyReply Direct Link To This Post Posted: 10/04/2013 at 11:04pm
Here is a document from USATT that contains charts for various handicap games:

http://www.teamusa.org/~/media/USA_Table_Tennis/Documents/Tournament%20Guide/CH%2006%20-%20Draw%20Procedures.pdf

The charts start on page 6-7.  We have used these charts for handicap matches with random draws, they are a lot of fun.
Back to Top
benfb View Drop Down
Platinum Member
Platinum Member


Joined: 10/10/2008
Location: United States
Status: Offline
Points: 2709
Post Options Post Options   Thanks (0) Thanks(0)   Quote benfb Quote  Post ReplyReply Direct Link To This Post Posted: 10/04/2013 at 11:15pm
The problem with playing multiple games is the fact that each games ends and then you start another.  If you're the stronger playing and giving a handicap, you need time to overcome that handicap.  Splitting the scoring into multiple games makes that more difficult.

For example, suppose the lower player wins the first game by luck (11-9) and then loses the second game by a wide margin (11-4, assuming a 4 point handicap).  In the third game the match still stands at 1-1 and there is that same handicap for the next game.  If you're only playing 2/3 games, then the situation becomes more difficult for the upper player. By comparison, if it's all played as one game, then the score is now 20-21 (I'm assuming a 12 point handicap on a 33 point game) and the upper player should win.

I think was the reason the USATT recommended single game handicap systems.
Back to Top
tt4me View Drop Down
Gold Member
Gold Member
Avatar

Joined: 01/17/2013
Location: RC Poverty Zone
Status: Offline
Points: 1019
Post Options Post Options   Thanks (0) Thanks(0)   Quote tt4me Quote  Post ReplyReply Direct Link To This Post Posted: 10/04/2013 at 11:18pm
I see no mathematical justification for the USATT tables.  The USATT table tends to provide handicap balls with a smaller ratings difference.  I can understand the better player being under pressure as Benfb said if the USATT handicap are used.

Back to Top
Baal View Drop Down
Forum Moderator
Forum Moderator


Joined: 01/21/2010
Location: unknown
Status: Offline
Points: 14213
Post Options Post Options   Thanks (0) Thanks(0)   Quote Baal Quote  Post ReplyReply Direct Link To This Post Posted: 10/04/2013 at 11:43pm
I definitely think tt4me's system is an upgrade, since their is a rational justification for it.
Back to Top
benfb View Drop Down
Platinum Member
Platinum Member


Joined: 10/10/2008
Location: United States
Status: Offline
Points: 2709
Post Options Post Options   Thanks (0) Thanks(0)   Quote benfb Quote  Post ReplyReply Direct Link To This Post Posted: 10/05/2013 at 2:02am
I took a look through tt4me's math and it's neatly done.  That said, there are some points that still concern me:

As tt4me has already pointed out, a lot of this depends upon the accuracy and consistency of USATT ratings.  the problem is that we already know they're neither accurate nor consistent.  That's why the Ratings Central system was invented -- to give ratings some mathematical sense.  For example, I would never claim that a 100 point difference means the higher player should win 5 to 1 over the lower player.  Also, the importance of a 100 point difference is different at different rating levels.

The second problem is that not only is there a variance on the probability of winning, there's also a variance on the ratings themselves.  One thing this tells us is that you need a lot of points played to gain confidence in the match outcome.  Of course, we know this from even play, where playing 4 out of 7 produces a more reliable result than just 2 out of 3.  But with handicap systems, it becomes even more important.

However, I think what bothers me most is breaking the match up into multiple games.  I worry (but haven't tried to do the math) that this changes the calculations.  I believe it will make the handicapping system less accurate.  I know that in even play, the change from 21 point games to 11 point games was mainly for psychological reasons (people tending to have long "runs" in 21 point games) but in a handicap game, the tension of coming from behind lessens that risk.

Interestingly, when I play players far below me in normal club play, I often provide a handicap, but not in ratings.  I choose a particular serve (or maybe two) and use that same serve on every point in the match.  If that's not sufficient, then I use the same serve to the same exact location every time.  If that's not sufficient, then I just hit an easy top spin ball to the same location every time (allowing them to start a rally).  Of course, all those handicaps only work on my serve.  and if none of those are sufficient? Then it's time to just give them lessons on basic strokes. LOL
Back to Top
tt4me View Drop Down
Gold Member
Gold Member
Avatar

Joined: 01/17/2013
Location: RC Poverty Zone
Status: Offline
Points: 1019
Post Options Post Options   Thanks (0) Thanks(0)   Quote tt4me Quote  Post ReplyReply Direct Link To This Post Posted: 10/05/2013 at 3:47am
Originally posted by benfb benfb wrote:

I took a look through tt4me's math and it's neatly done.  That said, there are some points that still concern me:

As tt4me has already pointed out, a lot of this depends upon the accuracy and consistency of USATT ratings.  the problem is that we already know they're neither accurate nor consistent.  That's why the Ratings Central system was invented -- to give ratings some mathematical sense.  For example, I would never claim that a 100 point difference means the higher player should win 5 to 1 over the lower player.  Also, the importance of a 100 point difference is different at different rating levels.
Look at the USATT ratings table.   There is a ratings difference range of 88-112 points so 100 is right in the middle of it.   Ratio of the points won or lost is 20 to 4 or 5 to 1.   That is my justification.  Obviously the order of wins a losses would make a difference over 6 matches but over the long haul the better player better win 5 games for every loss.  Ratings Central has a more continuous function for figuring out how much the players win or lose but it is the same idea.  

Quote
The second problem is that not only is there a variance on the probability of winning, there's also a variance on the ratings themselves.
Like I said, the problem is with the inaccuracy of the ratings system, not the handicap table.  The handicap table is computed so the expected outcome is 0.5.  There will always be a variance.  You can't eliminate that.   What other method would you use?

Quote
  One thing this tells us is that you need a lot of points played to gain confidence in the match outcome.  Of course, we know this from even play, where playing 4 out of 7 produces a more reliable result than just 2 out of 3.  But with handicap systems, it becomes even more important.
The goal is not to reduce confidence levels.  The goal is to find a handicap where both players have a 50-50 chance to win.   The confidence level for choosing which hand has the ball when deciding who serves first is relatively large but the odd of picking the ball right is still 50-50.  I agree that playing 4 out of 7 games would reduce the variance but like I said, that isn't the goal.   I thought the idea of handicaps is to make the match competitive so there is a 50-50 chance of winning and handicaps are mostly for fun and unrated tournaments.

Quote
However, I think what bothers me most is breaking the match up into multiple games.  I worry (but haven't tried to do the math) that this changes the calculations.
Certainly this changes the calculations.   Like I said,  if you play one game to 31 or 51 increase in handicap can occur at smaller intervals in ratings difference.   However, I can compensate by simulating a handicap of 5.5 balls by having a handicap of 6 then 5 then 6 etc or maybe 5 then 6 then 5.   I haven't investigate that yet.  The order in how the handicap is awarded should make any difference the first 3 games but it would in any extra games.  Edit, now I think about it, it should be relatively easy to simulate a 5 1/3 or 5 and 2/3 handicap.  That would make the increments in handicaps by 1/3 a ball so the increments in the ratings differences can be smaller.


I think I have provided something that is better than what exists.  I will entertain suggestions.  I can build new tables in no time.

Another brute force and simpler way to do handicaps is to simply remember the handicap between each pair of players.   If the better player wins the match then increase the handicap by one for the next match.  If the weaker player wins the match then reduce the handicap by one.  This is stupid simple but it requires keeping track of a lot of player combinations.
 






Edited by tt4me - 10/05/2013 at 3:51am
Back to Top
igorponger View Drop Down
Platinum Member
Platinum Member
Avatar

Joined: 07/29/2006
Location: Everywhere
Status: Offline
Points: 2962
Post Options Post Options   Thanks (0) Thanks(0)   Quote igorponger Quote  Post ReplyReply Direct Link To This Post Posted: 04/04/2017 at 10:53pm
THE BRITISH HAD MADE ALL PEOPLE EQUAL ... SO NICE OF THEM.

1.   An interesting compensative rule of the game, never known before.
Here is a rare vintage photo, presumably taken in the London Qeen's Hall, "free-for-all" money tournament, ca. 1902.



Two sporty-looking gentlemen at play.   You can clearly see a pronounced disparity in body size, -- the paddler positioned at the table's right end would top his opponent by a head or something.
IMPORTANTE OBSERVATION: The net displaced some 0,5 ft from the table's centre line towards the shorty player.   
Can you tell something more about this unusual handicapping (the net displacement)?   Is it still in use someplace?   
2. Another funny handicapping I would commend so much for club play is NOT to use your master-hand when playing a newcomer or a lower ranked playmate. Be happy...
   
------------
In truth .. The God made all people big and small, and Mr.Colt made all people equal.   And the Britishers made all the gamers equal as well. [:D:] [:D:]

    Handicapping in sports and games is the practice of delivering advantage through scoring compensation or other advantage given to different contestants to equalize the chances of winning.

    A handicap is a way to enable a weaker player to have a chance of winning against a stronger one. There are a variety of such handicaps, such as ...
Back to Top
 Post Reply Post Reply
  Share Topic   

Forum Jump Forum Permissions View Drop Down

Forum Software by Web Wiz Forums® version 12.01
Copyright ©2001-2018 Web Wiz Ltd.

This page was generated in 0.203 seconds.

Become a Fan on Facebook Follow us on Twitter Web Wiz News
Forum Home | Go to the Forums | Forum Help | Disclaimer

MyTableTennis.NET is the trading name of Alex Table Tennis Ltd.

Copyright ©2003-2020 Alex Table Tennis Ltd. All rights reserved.