Article - ATP Break Back Percentages Based On Time Decay & Break Back Stats

Skype: @TennisRatings

For my next article, I decided to do a detailed analysis of break back percentages in the ATP Tour based two variables.  

Being able to estimate the percentage chances of a break back occurring in a set is a very valuable asset because it enables us to be on the right side of many swing trading opportunities.  Especially in the ATP, an initial break lead reduces a player's price in-play significantly, and a break back will return the price to a similar level to where the price was before the break occurred.

The two variables are the time decay of the set, and the break back stats of the two competing players.  The data was derived from July 1st, 2013 until October 15th, 2013.

The time decay of the set is measured by the amount of games expired immediately after the break lead occurs, and the break back stats are measured by adding the sum of the percentage that the relevant player loses a break lead in a set added to the percentage their opponent recovers a break deficit in a set (all stats derived from the Ultimate In-Play Spreadsheet).

The break back stats are then broken down into five categories:-

Sum of Under 50.00
Sum between 50.00 and 59.99
Sum between 60.00 and 74.99
Sum between 75.00 and 89.99
Sum Over 90.00

With the average percentage in 2013 for a top 100 player, according to the stats on the Ultimate In-Play Spreadsheet, giving a break lead up being 32.14%, and getting a break deficit back being 35.42% (this is higher due to the fact that they play players outside the top 100) the average sum can be considered to be 67.56% - almost the midpoint of the third, middle category.  We can also then assume that the average percentage in ATP matches for a break back to occur is 33.78% (67.56%/2).

As we can see based on the table further below (also available via the resource link here), the stats based on the combined sum of the two players show a huge trend towards the higher percentages.

When the two players combined score is under 50, there have been 22.49% of break backs, and this barely rises when the combined score is between 50 and 59.99, where there were 22.81%.  It is clear from this data that laying the player a break up in this scenario has a very low chance of success - 11.29% and 10.97% below average respectively.  

Whilst readers may think this is fairly obvious, and would occur when top players are a break up, it also occurs in a great variety of other scenarios.  For example the likes of Philipp Kohlschreiber and Janko Tipsarevic have atrocious stats for getting a break deficit back and many of their matches would have been included these two categories.

It's interesting to see that with the average break back percentage being 33.78% in the ATP, the break back stats for the middle bracket - 60.00 to 74.99 were almost exactly average (as mentioned previously, the midpoint is almost exactly the combined average) giving 34.66% of break backs.

The break back stats greatly increase when looking at the 75.00 to 89.99 and over 90 combined scores - with break backs occuring 40.63% and 43.75% respectively in these brackets.

We can also see from the table below that break backs occur at different percentages based on the time decay of the set (games completed).

An average player that breaks in either the first and second games of the set has a slightly above 40% chance of getting broken back - just under 7% above average - and that is unsurprisingly the highest percentage for expired time due to the player having more time to 'defend' his break lead.

However the percentage from game 3 to game 6 is fairly consistent and around average (36.98% dropping to 31.37%) so it can be considered that when the combined score is high and the break lead occurs on or before game 6 of the set, the break recovery will occur on average just under 50% of the time.

There are also circumstances where the break back will happen more often than 50% - which should provide us with great opportunities for laying the player a break up...

These occur when the combined score is over 75, and the break occurs in the first or second game of the set (e.g. for *1-0 or 2-0*).  We can see from the stats below in this scenario that the break back occurred 85 out of the 147 times sampled (57.82% of the time).

Going back to the percentages by time decay, we can see that from game 8 or later, the percentages for the break back drops markedly.  A break after game 8 of a set will generate a *5-3 scenario, and after game 9 will generate a *5-4 scenario, so the player who has just obtained a break lead will be serving for the set immediately.  It appears that laying the player a break up in these situations (just 10.75% and a 14.13% break back percentage respectively) is not a viable proposition...

This is because the percentages are well below the current ATP average for breaking opponents (21.5%) showing that players serving for the set hold more often than average.  Whilst I haven't performed this analysis for the WTA yet I can say that I have previously researched this area in the WTA and found similar below average figures.   Players getting broken serving for the set obviously does happen but because gamblers/traders remember these instances more vividly than an early lead because of the 'choke' factor and that it has a huge influence on their positions in the match, and especially due to the popularity of social media also discussing this phenomenon, the gambling public appears to think it happens more often than it actually does.  

It's worth noting the phenomenon of the 11th game of the set though.  17 break backs occurred in 65 instances (26.15%) of a player breaking for a *6-5 advantage.  

There could be several reasons for this - firstly the sample size isn't the biggest and I will add to all these stats when I have more data in the future.  Secondly it could be because the set is closer than other scenarios and will often feature players more evenly matched on that basis.  Finally it could also be because the server feels under more pressure to hold serve because it's a one shot chance of winning the set - at *5-3 or *5-4 for example he knows that even if he gets broken he will still have another shot at his opponent's serve.

Based on my research, it's clear to see that combining time decay and the break back data provided in the Ultimate In-Play Spreadsheet will generate many positive expectation opportunities for profit.  If you laid the player blindly a break up when the combined score was over 75, you'd have had 427 entry points in the 3 and a half months sampled, and had 176 winning trades (41.22%) and in the vast majority of those, your profit would be significantly higher than your potential loss.

The complete data can be seen below.


Break After Game x of set = A player leads by a break immediately after that game of the set
Combined Score = The sum of the % the relevant player loses a break lead in a set added to the % their opponent recovers a break deficit in a set (all stats derived from the Ultimate In-Play Spreadsheet).
The figure in red for each game of the set is the overall break back percentage for that set.

ATPStats July 1st 2013 Average =33.78%
to 15th OctoberSample Average =32.69
Break After Game 1 of setBreak After Game 2 of setBreak After Game 3 of setBreak After Game 4 of set
CombinedBack on Serve%CombinedBack on Serve%CombinedBack on Serve%CombinedBack on Serve%
Break After Game 5 of setBreak After Game 6 of setBreak After Game 7 of setBreak After Game 8 of set
CombinedBack on Serve%CombinedBack on Serve%CombinedBack on Serve%CombinedBack on Serve%
Break After Game 9 of setBreak After Game 11 of setOverall
CombinedBack on Serve%CombinedBack on Serve%CombinedBack on Serve%
A Selection of TennisRatings Products
Please visit the
TennisRatings Products links for a full overview of our fantastic Tennis Trading tools, and the TennisRatings Subscription Packages link to see our great value range of discounted subscription packages!

Please check out our testimonials page!

The TennisRatings Daily Trading Spreadsheets have never been more popular!  

To find out more on how these can dramatically improve your Tennis Trading, check out the YouTube Video we made.

The Challenger Daily Spreadsheets cover all ATP Challenger Events and include projected hold percentages (for traders) and model prices (for bettors and traders).

Subscriptions are available for 3 months:-

The Lead Loss/Recovery Data Spreadsheets have taken the Tennis Trading World by storm - discussed in detail in October 2015 at the Matchbook Traders Conference these incredible spreadsheets highlight lead loss & deficit recovery in individual sets, as well as how often a player loses/gains the first break of the second set based on whether they won or lost the first set!