Members Login
Username 
 
Password 
    Remember Me  
Post Info TOPIC: Week 8 - ATP Challenger 100 - Pune, India (hard)


Lower Club Player

Status: Offline
Posts: 126
Date:
RE: Week 8 - ATP Challenger 100 - Pune, India (hard)


JonH comes home wrote:
B00thy wrote:
JonH comes home wrote:

Let's hope he achieves what he deserves this week.

The 8 players left, though, arent slouches.

Not sure how the bookies see it but Tennis Abstract rate Billy as a bit of an outsider to take the title - Brandon Holt is up next and they rate it 59-41 in his favour to beat Billy. If Billy does come through, Vacherot is rated above him and Holt and Vacherot are the top 2 overall. Of the 8, Billy is rated a 10% chance of the title (so less than an even chance in an 8 players field) and 5th favourite overall.

You show them Billy!

Player SF F W
(1)Billy Harris (GBR) 40.6% 18.5% 10.2%
(6)Brandon Holt (USA) 59.4% 32.7% 21.0%
Valentin Vacherot (MON) 61.5% 32.8% 20.9%
(8)Alexis Galarneau (CAN) 38.5% 16.0% 8.4%
(5)Ugo Blanchet (FRA) 62.3% 33.0% 13.8%
Khumoyun Sultanov (UZB) 37.7% 15.5% 4.7%
Dalibor Svrcina (CZE) 71.3% 41.2% 18.5%
(Q)Ilia Simakin (RUS) 28.7% 10.3% 2.5%


 excuse my ignorance .. if Billy is an outsider to take the title then how did he get seeded 1? could explain how you think this came about? Thanks


 As I say, the bookies may see it differently. And as Indy says, there is another way of looking at it. Tennis Abstract, in terms of how they do it, use an algorithm they have created that is ELO. In essence, it takes the players recent form and who they have played and assigns an ELO rating that is then adjusted for surface; they then generate 1000's of simulations of the match and the % shown is the output of that simulation exercise. 

Billy has an ELO rating of 1597 currently, which places him 168 in the TA view of the current rankings; as opposed to his real ranking of 110 or whatever it is that gave him the seeding. They assess Holt has an ELO of 1653 and a ranking of 134, which is better than his actual world ranking (and why he has a lower seeding). 

They put these ELO's together, adjust for surface in some way, and then run the simulations. They do that for every match in the draw, through each round in terms of each potential match up and work out round by round percentages and an overall view.

In Glasgow, for example, Evo is rated as tournie favourite at 28.4% as it stands; Jake is also rated favourite in Pau at 43.1% - the percentages change as every match finishes and is updated as a player winning a match takes another out of the draw and changes the relative positions of each player and their potential progress through the draw - so if you check these percentages in an hour or two, they could vary if the field has changed

https://www.tennisabstract.com/current/2025PuneChallenger.html

https://www.tennisabstract.com/current/2025GlasgowChallenger.html

https://www.tennisabstract.com/current/2025PauChallenger.html

https://www.tennisabstract.com/current/2025ATPDoha.html

In Dubai, Jack is currently third favourite of the QF field left. Alcaraz is the favourite, obviously

https://www.tennisabstract.com/ is the main site link and you can see lots of other stuff there as well as womens draws and analysis. They only analyse it for ATP/WTA main tour and Challenger/125 events though, not ITF

 This explains how ELO ratings work and how it flows through to predictions at a match level

https://tennisabstract.com/reports/atp_elo_ratings.html

A 100-point difference in Elo ratings implies that the favorite has a 64% chance of winning a best-of-three-set match; 200 points implies 76%, 300 points implies 85%, 400 points implies 91%, and 500 points implies 95%. In best-of-five, the favorite is more likely to win, by a factor that depends on the best-of-three odds.

Surface-specific Elos--"hElo," "cElo,", and "gElo"--are a mix of overall Elo and separate ratings generated using only results on the given surface. These ratings give more accurate forecasts for individual matches.   



-- Edited by JonH comes home on Thursday 20th of February 2025 11:48:33 AM


 Thanks  for the explanation - illuminating! so essentially the seeding is on ranking but the top ranked player isn't necessarily in the same form as others 



__________________


Tennis legend

Status: Offline
Posts: 43253
Date:

B00thy wrote:
JonH comes home wrote:
B00thy wrote:
JonH comes home wrote:

Let's hope he achieves what he deserves this week.

The 8 players left, though, arent slouches.

Not sure how the bookies see it but Tennis Abstract rate Billy as a bit of an outsider to take the title - Brandon Holt is up next and they rate it 59-41 in his favour to beat Billy. If Billy does come through, Vacherot is rated above him and Holt and Vacherot are the top 2 overall. Of the 8, Billy is rated a 10% chance of the title (so less than an even chance in an 8 players field) and 5th favourite overall.

You show them Billy!

Player SF F W
(1)Billy Harris (GBR) 40.6% 18.5% 10.2%
(6)Brandon Holt (USA) 59.4% 32.7% 21.0%
Valentin Vacherot (MON) 61.5% 32.8% 20.9%
(8)Alexis Galarneau (CAN) 38.5% 16.0% 8.4%
(5)Ugo Blanchet (FRA) 62.3% 33.0% 13.8%
Khumoyun Sultanov (UZB) 37.7% 15.5% 4.7%
Dalibor Svrcina (CZE) 71.3% 41.2% 18.5%
(Q)Ilia Simakin (RUS) 28.7% 10.3% 2.5%


 excuse my ignorance .. if Billy is an outsider to take the title then how did he get seeded 1? could explain how you think this came about? Thanks


 As I say, the bookies may see it differently. And as Indy says, there is another way of looking at it. Tennis Abstract, in terms of how they do it, use an algorithm they have created that is ELO. In essence, it takes the players recent form and who they have played and assigns an ELO rating that is then adjusted for surface; they then generate 1000's of simulations of the match and the % shown is the output of that simulation exercise. 

Billy has an ELO rating of 1597 currently, which places him 168 in the TA view of the current rankings; as opposed to his real ranking of 110 or whatever it is that gave him the seeding. They assess Holt has an ELO of 1653 and a ranking of 134, which is better than his actual world ranking (and why he has a lower seeding). 

They put these ELO's together, adjust for surface in some way, and then run the simulations. They do that for every match in the draw, through each round in terms of each potential match up and work out round by round percentages and an overall view.

In Glasgow, for example, Evo is rated as tournie favourite at 28.4% as it stands; Jake is also rated favourite in Pau at 43.1% - the percentages change as every match finishes and is updated as a player winning a match takes another out of the draw and changes the relative positions of each player and their potential progress through the draw - so if you check these percentages in an hour or two, they could vary if the field has changed

https://www.tennisabstract.com/current/2025PuneChallenger.html

https://www.tennisabstract.com/current/2025GlasgowChallenger.html

https://www.tennisabstract.com/current/2025PauChallenger.html

https://www.tennisabstract.com/current/2025ATPDoha.html

In Dubai, Jack is currently third favourite of the QF field left. Alcaraz is the favourite, obviously

https://www.tennisabstract.com/ is the main site link and you can see lots of other stuff there as well as womens draws and analysis. They only analyse it for ATP/WTA main tour and Challenger/125 events though, not ITF

 This explains how ELO ratings work and how it flows through to predictions at a match level

https://tennisabstract.com/reports/atp_elo_ratings.html

A 100-point difference in Elo ratings implies that the favorite has a 64% chance of winning a best-of-three-set match; 200 points implies 76%, 300 points implies 85%, 400 points implies 91%, and 500 points implies 95%. In best-of-five, the favorite is more likely to win, by a factor that depends on the best-of-three odds.

Surface-specific Elos--"hElo," "cElo,", and "gElo"--are a mix of overall Elo and separate ratings generated using only results on the given surface. These ratings give more accurate forecasts for individual matches.   



-- Edited by JonH comes home on Thursday 20th of February 2025 11:48:33 AM


 Thanks  for the explanation - illuminating! so essentially the seeding is on ranking but the top ranked player isn't necessarily in the same form as others 


 Nice succinct summary! 



__________________


Tennis legend

Status: Offline
Posts: 43253
Date:

And of course it isnt always right! Evo was rated clear favourite by the TA ELO based algorithm and yet lost in straight sets. Thankfully it isnt perfect otherwise theres no point playing, just let the computer anoint a winner!

__________________


Intermediate Club Player

Status: Offline
Posts: 367
Date:

Indeed. Quite interesting and sensible to ignore more the churlish contributors.

__________________


Tennis legend

Status: Offline
Posts: 43253
Date:

After a poor start where he went 0-3, Billy did well to win that first set 6-4. Come on Billy, prove those Tennis Abstract algorithms wrong!

__________________


Tennis legend

Status: Offline
Posts: 43253
Date:

And a break set two now, serving at 3-2

__________________


Hall of fame

Status: Offline
Posts: 9463
Date:

Then lost the next 4 games...one set all

__________________


Tennis legend

Status: Offline
Posts: 43253
Date:

Not able to watch at all, just following scores, but it seems a very up and down match - one goes on a burst then the other. Billys now got a third set break but suspect more twists to come.

__________________


Lower Club Player

Status: Offline
Posts: 126
Date:

JonH comes home wrote:
B00thy wrote:
JonH comes home wrote:
B00thy wrote:
JonH comes home wrote:

Let's hope he achieves what he deserves this week.

The 8 players left, though, arent slouches.

Not sure how the bookies see it but Tennis Abstract rate Billy as a bit of an outsider to take the title - Brandon Holt is up next and they rate it 59-41 in his favour to beat Billy. If Billy does come through, Vacherot is rated above him and Holt and Vacherot are the top 2 overall. Of the 8, Billy is rated a 10% chance of the title (so less than an even chance in an 8 players field) and 5th favourite overall.

You show them Billy!

Player SF F W
(1)Billy Harris (GBR) 40.6% 18.5% 10.2%
(6)Brandon Holt (USA) 59.4% 32.7% 21.0%
Valentin Vacherot (MON) 61.5% 32.8% 20.9%
(8)Alexis Galarneau (CAN) 38.5% 16.0% 8.4%
(5)Ugo Blanchet (FRA) 62.3% 33.0% 13.8%
Khumoyun Sultanov (UZB) 37.7% 15.5% 4.7%
Dalibor Svrcina (CZE) 71.3% 41.2% 18.5%
(Q)Ilia Simakin (RUS) 28.7% 10.3% 2.5%


 excuse my ignorance .. if Billy is an outsider to take the title then how did he get seeded 1? could explain how you think this came about? Thanks


 As I say, the bookies may see it differently. And as Indy says, there is another way of looking at it. Tennis Abstract, in terms of how they do it, use an algorithm they have created that is ELO. In essence, it takes the players recent form and who they have played and assigns an ELO rating that is then adjusted for surface; they then generate 1000's of simulations of the match and the % shown is the output of that simulation exercise. 

Billy has an ELO rating of 1597 currently, which places him 168 in the TA view of the current rankings; as opposed to his real ranking of 110 or whatever it is that gave him the seeding. They assess Holt has an ELO of 1653 and a ranking of 134, which is better than his actual world ranking (and why he has a lower seeding). 

They put these ELO's together, adjust for surface in some way, and then run the simulations. They do that for every match in the draw, through each round in terms of each potential match up and work out round by round percentages and an overall view.

In Glasgow, for example, Evo is rated as tournie favourite at 28.4% as it stands; Jake is also rated favourite in Pau at 43.1% - the percentages change as every match finishes and is updated as a player winning a match takes another out of the draw and changes the relative positions of each player and their potential progress through the draw - so if you check these percentages in an hour or two, they could vary if the field has changed

https://www.tennisabstract.com/current/2025PuneChallenger.html

https://www.tennisabstract.com/current/2025GlasgowChallenger.html

https://www.tennisabstract.com/current/2025PauChallenger.html

https://www.tennisabstract.com/current/2025ATPDoha.html

In Dubai, Jack is currently third favourite of the QF field left. Alcaraz is the favourite, obviously

https://www.tennisabstract.com/ is the main site link and you can see lots of other stuff there as well as womens draws and analysis. They only analyse it for ATP/WTA main tour and Challenger/125 events though, not ITF

 This explains how ELO ratings work and how it flows through to predictions at a match level

https://tennisabstract.com/reports/atp_elo_ratings.html

A 100-point difference in Elo ratings implies that the favorite has a 64% chance of winning a best-of-three-set match; 200 points implies 76%, 300 points implies 85%, 400 points implies 91%, and 500 points implies 95%. In best-of-five, the favorite is more likely to win, by a factor that depends on the best-of-three odds.

Surface-specific Elos--"hElo," "cElo,", and "gElo"--are a mix of overall Elo and separate ratings generated using only results on the given surface. These ratings give more accurate forecasts for individual matches.   



-- Edited by JonH comes home on Thursday 20th of February 2025 11:48:33 AM


 Thanks  for the explanation - illuminating! so essentially the seeding is on ranking but the top ranked player isn't necessarily in the same form as others 


 Nice succinct summary! 


 I've been thinking about this quite a bit .. I have tended to unthinkingly assume that the seeds are the best way of considering the strength of forthcoming match I'm interested in. Not any more. Someone dropping down the rankings through loss of form will still be seeded according to their ranking whereas I am interested in recent form going into a match so I will be exploring this subject a little more. Thanks again JonH.



__________________


Tennis legend

Status: Offline
Posts: 43253
Date:

Cool - no worries. As I say, it is fallible, Evo for example was rated tournament favourite for Glasgow but we all saw what happened there and , being frank, he cant buy a win at the moment.

But certainly the real rankings dont provide the best guide, some sort of form related guide is likely to be better. Whether Tennis Abstract is the best around, I dont know. Its there, its accessible and the person who runs it, Jeff Sackmann, is an acknowledged tennis stats leader.

But apps like TNNS have win predictions built into their matchups and I have no idea where they take their data from, maybe it is Tennis Abstract or some other method - Ive never compared their percentages so dont know. Im sure the bookies also take similar feeds into account in their odds setting as well as other risk related factors

__________________


Tennis legend

Status: Offline
Posts: 43253
Date:

Meanwhile Billy down 4-5 and a break final set.

__________________


Lower Club Player

Status: Offline
Posts: 126
Date:

Holt playing deep shots in the last service game of Billy's pressurising Billy into errors



-- Edited by B00thy on Friday 21st of February 2025 09:10:19 AM

__________________


Tennis legend

Status: Offline
Posts: 43253
Date:

And Billy goes down, thats a real shame but not a massive shock ( see above!)

Not sure how Billy will look back on this Indian tour? Played 3 events, 8 wins and 3 losses, a final, semi and quarter.

He didnt get the title Im sure he craves nor get to break the top 100 barrier. But Id say that his form improved on recent times and he should probably feel hes got some momentum to build on.

__________________


Tennis legend

Status: Offline
Posts: 41322
Date:

Ah pity. Close one.

But yes, overall pretty decent form from Billy with 3 months to come in which he is still due to drop relatively few ranking points.

He ends his India run live ranked 109 on 534 points, 48 points behind the current ( and live ) WR 100 on 582 points.

He just has a total of 38 points ( net 17 ) from 3 counters due to drop up to and including the pre French Open / grass season 26/05 ranking date.

In the immediate future he has understandably pulled out of next week's Accapulco. Then he is down for Indian Wells qualifying, with presumably Miami to follow, so likely tough but with good points available even just for qualifying. Then, if still short of that top 100, pick and choose wisely over the following 8 weeks. Asia hard court challengers maybe, not that he is too bad on clay?

It looks very doable. Hope so!



-- Edited by indiana on Friday 21st of February 2025 10:23:23 AM

__________________


Tennis legend

Status: Offline
Posts: 43253
Date:

indiana wrote:

Yes, overall pretty decent form from Billy with 3 months to come in which he is still due to drop relatively few ranking points.

He ends his India run live ranked 109 on 534 points, 48 points behind the current ( and live ) WR 100 on 582 points.

He just has a total of 38 points ( net 17 ) from 3 counters due to drop up to and including the pre French Open / grass season 26/05 ranking date.

In the immediate future he has understandably pulled out of next week's Accapulco. Then he is down for Indian Wells qualifying, with presumably Miami to follow, so likely tough but with good points available even just for qualifying. Then, if still short of that top 100, pick and choose wisely over the following 8 weeks. Asia hard court challengers maybe, not that he is too bad on clay?

It looks very doable. Hope so!


 I think that is a good assessment - doable, but needs to pick wisely, both for the top 100 points he needs and I am sure he would love to get a title in there somewhere along the way. Feels to me Asia and hardcourts would be the wise choice, one thing we do know is Billy will work hard and he WILL play lots of tournies 

Out of interest, what is his lowest counter he is trying to improve on and what does that basically align to in terms of challenger or ATP events (ie what round would he need to reach to improve his total?) 



__________________
«First  <  1 2 3 4  >  Last»  | Page of 4  sorted by
 
Quick Reply

Please log in to post quick replies.

Tweet this page Post to Digg Post to Del.icio.us


Create your own FREE Forum
Report Abuse
Powered by ActiveBoard