|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| >> Current Period | | | Hard Court Swing | | | Aug 1 2008 to Sep 7 2008 | |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
|
TennisInsight.com offers the most comprehensive statistical analysis of men’s tennis you’ll find on the web, featuring in-depth previews of every ATP tennis match on the Men’s Tour. Our detailed match records, head-to-head results and key player statistics are all underpinned by the Insight Tennis Model that statistically rates every ATP player and every match.
The statistics presented here at TennisInsight.com may be a little daunting at first, but if you take a few moments to read the remainder of this section you will be equipped with everything you need to get the most out of TennisInsight.com and become a guru of the Men’s Tour.
| Finding your way around TennisInsight.com | |
All upcoming matches are displayed on the shortcut panel on the right hand side of screen, whilst Player Profiles can be viewed at any time by clicking on the player’s name (or using the search function at the top of screen). You can also create your own H2H player comparisons via the Match Previews section, or within a player profile. After a short time on the site you should find the navigation quick and easy.
You will notice that player profiles, match previews and custom player comparisons all have the same structure of overview, activity, highlights and stats. The following example compares Roger Federer with Andre Agassi to give you an idea of the content on these pages. Note that at any time you can toggle between each player’s info by simply clicking on their names. Check out the notes in the bottom section of each page for more hints about using TennisInsight.com.
Click on each of the following for examples comparing Federer and Agassi:
>> Overview
>> Activity
>> Highlights
>> Stats.
| Understanding Ratings and Handle | |
Who is going to win? Will it be a close Match? How do the players compare? As any regular tennis fan knows the ATP Rankings and Seedings are often a poor good guide to answering these sorts of questions. What we need is an evaluation of form – assessments of just how well each player is playing – and that's where the Insight Tennis Model comes into play.
The Insight Tennis Model combines fundamental statistical techniques with a deep knowledge of tennis to produce a sophisticated analysis of player form. In line with any fundamental statistical investigation the outputs of the model describe the mean (Ratings) and the variation (Handle) as described below.
The Insight Tennis Model assigns numerical ratings to each player based on a statistical assessment of their match results over time. The ratings typically range anywhere from -5 up to +5 or even higher. The rating values have no inherent meaning, however when Comparing Players the difference in Match Rating between the two players represents the average number of Service Games that separate them in a set of tennis. Note that this is a long-term average prediction so for example if we rated Rafael Nadal 2 games better than his opponent we would expect that as more sets of tennis were played the average score line will tend toward a 2-game advantage to Nadal.
| >> Match Rating versus Surface Rating | |
As you browse through TennisInsight.com you will notice that both Match Ratings and Surface Ratings are displayed – so what's the difference? Simply put, Surface Ratings assess the underlying form of a player for the given playing surface. Factors such as win-loss record and the quality of previous opposition are taken into account when calculating Surface Ratings. Alternatively Match Ratings assess form specific to the current tournament conditions and the opposing player (Match Ratings now take H2H into account).
For every single match result the Match Rating of each player is displayed in brackets immediately after the players name (along with the players rank at the time of the match).
| >> Getting a handle on it! | |
Assigning a statistical rating to each Player is easy with the Insight Tennis Model. But how much credibility can we give to that Rating? As stated above the Ratings are assessed based on match results over time, hence the more Matches played and the more recently those matches were played, the greater handle we have on our statistical assessment.
Equivalent to Confidence Interval, Handle values on TennisInsight.com range from zero ("no handle") to 1 ("as much handle as can be") with values less than 0.5 considered poor. In this case the predicted form rating is considered speculative and the player's 'true' form may be significantly different from the Insight Tennis Model rating.
In our Match Previews we display the highest Surface rating for the given surface ever achieved for that player. In this instance these ratings are tied a handle value of greater than 0.5.
| >> Predicted Chance and Minimum Buy-In | |
Based on the Ratings and Handles for each player, a combined statistic is calculated that gives an estimate of the chance of either player to win the match. This is shown as a range (a confidence interval) – the better Handle we have on each player, the smaller the range.
The minimum buy-in is simply the reciprocal of the Models minimum estimated chance for each player and represents the lowest price at which you would bet if you went purely off the Model Rating.
| Other Statistics at TennisInsight.com | |
In the Activity and Stats sections of a Player Profile or Match Preview you will notice 'Avg X' and 'Avg Delta', here is everything you need to know to interpret these stats:
The Insight Tennis Model assigns a statistically rated score (X-score) for every player for every match. These scores form the basis for calculating the Ratings discussed above and they are also presented on the site to give further insight into how the players are performing.
“Avg x” is a straight-line average of a given set of x-scores. This simple value can be particularly useful when looking at how a player typically performs in certain conditions - for example on a given surface, versus top 10 players, or in any of the conditions that may be selected in the activity filter.
“Avg  ” is simply the difference between actual and predicted performance. Players with a positive Avg are typically outperforming the the Insight Tennis Model's expectations for the given set of matches.
In the Tournament Profiles section you will notice we display a statistical assessment of court speed. Generally, matches on faster courts have fewer points, more games and more tie breakers. "Court speed" is thus inferred from completed match results and whilst not a true measure of how quickly the court is playing it does provide some useful insight on this. (Click here for an example.)
| >> Updating Match Ratings | |
There was a time when we displayed retrospectively updated ratings after the match. This is no longer the case. In other words, when you browse through completed matches, the Ratings displayed are the predicted ratings determined ahead of the match.
TennisInsight.com contains all Main Tour and Challenger results (including qualifying) along with certain exhibition results dating back to January 2000. Special thanks to SteveG for providing these historical results.
| Frequently Asked Questions | |
We’ve put this FAQ together in response to a number of emails we have received about the site. If you have any additional questions, feel free to contact us.
No. There are plenty of sites out there that offer betting picks for tennis and some of these are very good. However following others tennis picks is a tough pastime in our view. For starters it’s hard to find the same prices as quoted by tipsters as these are often long gone by the time of publishing (or very soon after). Also, picks are often published only a short time before the days play commences, leaving you a very small window in which to respond. It’s also worth noting that regardless of whether you win or lose, the tipster has made money – your money.
Our view is that if you are going to bet on tennis then why not garner yourself with a little knowledge (TennisInsight.com is a good place to start!) and make your own betting decisions. This adds a whole lot of fun to the process and you never know, you may turn out to be one of the minority that have what it takes to make money out of tennis punting over the long term.
| >> What about your 'Minimum Buy-in'? | |
The minimum buy-in is simply the reciprocal of the Models minimum estimated chance for each player and represents the lowest price at which you would bet if you went purely off the Model Rating.
| >> How can I use TennisInsight.com to get more out of my punting? | |
The statistical Match Ratings and Minimum Buy-ins are starting guides for evaluating a particular match – but betting purely on the basis of the ratings is not recommended. The following points explain how we think you can use the stats here at TennisInsight.com to get more out of your punting experience:
>> Handle Level. As explained in the guide, anything less than 0.5 is considered very poor and reflects the fact that the player has not played very much tennis recently (either on that surface, or at all). However, this instance can provide very good betting opportunities as the bookies also have less info to go on in pricing a match, which leaves more room for error. Any specific knowledge about the player can be used to good effect in these scenarios.
>> Last 10 Delta. Displayed in the activity section, this shows how the player has been performing against the model ratings in recent times. Significant values either side of zero indicate that the model is not keeping up with the players change in form – and in this case it’s fair to say the Model Match rating is not worth blindly following.
>> Max Ratings. Compared to the current Match Rating, this can show where in the cycle a player is at. For example a young player may be consistently breaking into new highs (be cautious in opposing), or they could be coming back from injury and finding form – in this case the Max Rating can give an idea of what level of form they can reasonably expect to achieve again. Alternatively senior players playing near their highs can reasonably be expected to fall back a notch, rather than break into new levels of play.
>> H2H Results. Using the Model Ratings can show when players have a particular like or dislike for each others style of play.
>> Win-Loss record by Level, by Region. Displayed in the stats section, these illustrate where players have a particular like or dislike for certain situations.
>> Overall we recommend the same approach to punting as used by all good investors, which is a combination of technical (quantitative) and fundamental (qualitative) analysis. Examples of qualitative factors worth considering are injuries, physical conditioning, motivation, playing in the doubles draw and so on.
| >> How do you classify a big server? | |
“Big Servers” is one of the filter options when looking at player activity. Without actually knowing real data of player serve speeds, we use a simple statistical assessment based on the average length of sets played. This is based on the premise that big serving players are more likely to be involved in longer sets, which is clearly not always the case. We like to think of this stat as showing players with a high dependency on their serve. They don’t get broken often, but also do not break serve often and hence have a reliance on serving effectively to win matches.
| >> How do you derive “Max Ratings” for each surface? | |
Maximum Ratings are updated weekly and represent the highest ever Rating achieved on a given surface where the Handle is 0.5 or higher.
| >> What do you mean by ‘Handle’ ratings? | |
This has nothing to do with the player's racket! As per the explanation in the guide, this is an equivalent to ‘Confidence Interval’ which describes how valid the Model Rating is.
'Your Insight' lets you share your expert tennis knowledge with the rest of the TennisInsight.com community. Whilst there are plenty of chat rooms devoted to tennis, Your Insight is intended to be a little different and indeed far more useful. Our aim for Your Insight is to let tennis experts share useful knowledge in a contextual manner.
In order to achieve this aim, we ask you to make Your Insight contributions useful, contextual, and constructive.
Examples of useful information include:
- player quotes;
- information about player fitness, fatigue and injuries;
- observations about recent performances (or previous H2H performances) such as match stats or anything noteworthy;
- style of play and how it matches up to the opponent and/or the conditions;
- state of mind, motivation for the match;
- your analysis or preview of the match;
- your post-match assessment (you can add Your Insight to any historical match in our database);
- your analysis of the Insight Tennis Model prediction and why you think it may be valid or flawed; and
- any other information your fellow tennis enthusiasts will benefit from!
Your Insight posts are tied to a specific match. This contextual information means that as you navigate through recaps of historical matches you will be able to see comments that recreate the context under which those matches were played – which we hope will provide profound insight and learning to assist your tennis knowledge and punting acumen.
As statistical modellers and part-time gamblers we appreciate just how difficult it can be to predict tennis matches – so remember that while it is all too easy to criticise, we kindly ask you to make any criticism constructive!
| >> What is the future of Your Insight? | |
Our vision is simple – we strive to make TennisInsight.com the most useful resource for tennis enthusiasts (not just you punters!) on the web. Your Insight is an opportunity for you to help us achieve this vision and in return we pledge simply that as we grow and succeed we will reward all of those that contribute positively to this success with nothing short of cold hard cash (stay tuned)!
|
|
|
|
|
|
UlyssesWeb.com | SteveGTennis.com | OnCourt | Tennis Betting | Disclaimer | Guide | FAQ | About Us | Contact
|
|
|