IN-GAME TRADES 1 – BACKING THE GOOD SERVER

IN-GAME TRADES 1 – BACKING THE GOOD SERVER

As I previously mentioned, I’m not a fan of backing ‘good servers’, especially notoriously good servers in their service games due to the poor risk/reward ratio. However, one way that we can reduce the risk is to back players with a high projected hold (again – especially those players who aren’t known good servers) when losing in their service games.

Because we are backing the server when losing in their service game, the price we back the player at is higher than at a game score of 0-0. Therefore this will reduce our tick loss should the server lose their service game. Not only this, but our potential tick gain is greater because we are backing them at a better price than at the start of the service game.

Before this trading strategy begins to sound too good to be true, I need to point out something pretty vital. In-game liquidity isn’t the best. Personally I only trade in-game on live streamed matches only with £100,000 or more matched in the market before the match starts. For non-live matches, it’s possible to work out the in-game score in advance of a live score website by just watching the markets move but I would definitely not recommend this move for a beginner. With liquidity not being the best, it’s often difficult to get large trades matched too, so this type of trading strategy isn’t very scalable for non-live matches.

To cope with illiquid markets, you will need to offer prices. The more experienced you get in the markets, the easier you will find this as you will begin to be able to anticipate the tick movement per point depending on the match state. To give a very general example, if a player is something like 1.52 to back and 1.57 to lay immediately following a point, if you offer 1.54 or 1.55 either way you will have a very good chance of being matched.

You need to try and get the trade offered immediately after a point is finished because like tiebreaks, it’s vital to be very quick to get your trades matched. As soon as the server throws the ball in the air on the Betfair video, if my entry trade isn’t matched, I cancel it.

With the new legislation allowing just 20-25 seconds per point, you need to be very quick, especially considering Betfair have the 5 second delay for trades to be placed. This type of quick trading isn’t for everyone.

Now that’s out of the way, I wanted to discuss the various entry points for this strategy.

When I first decided to look at this trading strategy, I backtested various scenarios to see which entry points were more lucrative than others and achieved some pretty interesting results…

To start with, I ruled out any score that meant the server was down by one point (e.g. 0-15, 15-30, 30-40 or 40-A) because the tick loss will still be very heavy should the server be broken.

So that narrowed down the sample to 0-30, 0-40, or 15-40 as potential entry points. As I am a fairly risk-averse trader, I wanted to assess how my selected players would fare winning both the next point, and the next two points. I didn’t want to leave any winning trades any longer than that.

In the ATP, the average player across all surfaces wins 63.1% of service points (holding 78.2% of the time). I looked at players with a projected hold of over 81% (so at least 2.8% above ATP mean).

At 0-30, players with a projected hold of over 81% won the next point 66.7% of the time, well above those averages. They won the next two points on 46.7% of occasions (6.9% above the all surfaces average of 39.8% (63.1%*63.1%) so clearly this would be a very viable entry point as a big server wins significantly more points at this scoreline than would be expected of an average player.

However, at 0-40, the situation was vastly different. This was not a good entry point whatsoever. My results show that even strong servers struggle at this point – perhaps tennis players consider that game to be a foregone conclusion by that point. In my sample of strong servers, they only won the next point 48.8% of the time – well below the average for even the weakest of servers on any ‘normal points’. The stats for winning the next two points were also not fantastic – a success rate of 33.3%, again below average. So whilst the risk is incredibly low backing a server at 0-40, the reward is much less likely to happen than at 0-30.

Finally, I examined the stats for the 15-40 entry point. This actually was the best entry point of the three sampled. The server won 69.5% of the next service points, which is 6.4% above the fair average. Furthermore, the server won the next two points a huge 13.9% of the time. This is fantastic because if we enter at 15-40 with our exit points either at a break of serve or ‘greening up’ at 40-40 this means we are slightly more likely to win our trade than lose it – and there is no doubt that our tick profit will be greater going from 15-40 to 40-40 than it would be from 15-40 to service break so our winning trades will also have more profit than the losses our losing trades incur.

I also did the same sample for the WTA – looking at players which had projected holds over 67%. Remarkably, the results were almost parallel to the ATP.

As with the ATP, 0-40 was by far the worst entry point. The average WTA player wins 55.8% of their service points but in this sample women with high projected holds won just 55.0% of the time to take it to 15-40 – slightly below WTA average. However, when getting the next point to 15-40, the women often took the next point as well. Going from 0-40 to 30-40 occurred 38.1% of the time – above the 31.1% expectancy. So this would indicate that 15-

40 would be a viable entry point…

This was confirmed in emphatic style as in the sample the high projected hold women won the next point from 15-40 to 30-40 64.6% of the time (8.8% above WTA mean) and 37.5% won the next two points to take it to 40-40 (6.4% above the 31.1% expectancy). Clearly this is a very good entry point.

That can also be said for the 0-30 entry point where 60.8% of women with high projected holds won the next point to 15-30 in my sample – 5.0% above WTA mean. Not only this, they were very successful winning the next two points in my sample, to 30-30. This occurred on 42.9% of occasions – a huge 11.8% above average. So the statistics found this to be an excellent entry point as well.

It can therefore be said that 0-30 and 15-40 are excellent entry points to back players with high projected holds for both the ATP and WTA. These entry points have much less risk than backing these players at 0-0 in the game, and they produce significant rewards when they are successful.