Every Second Counts

As we’ve stated before, the Betting School here on BetPreviews.com is all about interactivity – you ask the questions, and we do our best to answer them. Sam Ryall got in touch after our previous post about special offers and here were his questions:

One bookie offers money back on second to the favourite in any race of 5 runners or more, with an SP of 3/1 or greater. Surely if you isolated all 5 runner races with odds of roughly 1/3 Fav, 3/1 2F, and 20/1 bar you would be guaranteed to win over time by backing the 2nd fav?

In general you would do well also by backing any horse in a field hovering at just above the 3/1 mark where there is also an odds on fav??

The first thing that needs to be clarified here is whether or not the bookie is refunding on any race where the favourite wins and the 2nd placed horse is 3/1 or more, or whether the favourite also has to be 3/1 or bigger. if it’s the latter, there’s not a lot of value in this under any circumstance, certainly not enough to compensate for the over-round.

However let’s take a theoretical race where as you say it’s 1/3 Apple, 10/3 Banana, 16/1 Carrot, 25/1 Damson, 33/1 Elderberry, and the special offer applies. The over-round here is 11%, which is reasonably accurate for an SIS showing. If we presume that the pricing is an accurate reflection of the chances of winning the race, then the true odds of winning are approximately 68% A, 20% B, 5.5% C, 3.5% D, 3% E. Thus, the straight forecast chances might be (strictly estimated, the CSF is one of the most complicated formulae in existence!) AB 42%, AC 12%, AD 8%, AE 6%. Note here that you want consistency here – avoid novice or maiden chases where the prospects of falling are high.

In that instance, your €10 bet on Banana at 10/3 now has a 20% chance of winning (expected value €8.66 approx) and a 42% chance of returning your €10 (expected value €4.20) meaning that you have got a huge edge here, of the order of 28.6%.

The critical aspect to all this is of course the straight forecast price. Essentially, if the percentage chance of your forecast in in excess of the percentage of the over-round, then that difference is the edge you have. For example, take today’s 4.10 at Haydock Park.  The Computer Straight Forecast, which is biased towards favourites and writes in a big margin for the bookie, suggested that there was maybe a true 20% chance of Humungosaur, the 9/2 second favourite, following home Calaf, the 8/13 favourite.  If so, then in this instance, then Humungosaur would have represented good betting value, possibly offering a 6-8% edge to the punter.

Also, a bookie offers cash back if your horse is second to any horse in virtual racing. I don’t know what kind of over-rounds virtuals work off but I presume its about 120%. Surely backing any virtual horse between even money and 6/4 would mean you would finish up over time by virtue of the refunds you would get?

Based on a very small sample across five bookmakers from today, virtual races seem to have an over-round of between 15% and 18%. Thus, you need to be sure that you have at least this much of a chance of a refund, in order for the numbers to stack up. The problem here is that bookmakers are cognisant of the fact that while they can do nothing about skewed books in actual horse racing, where the win odds allied to the each way terms can give value to the punter, they are hardly going to allow it in virtual racing. Thus a virtual race with an even money favourite and another short priced horse is unlikely to come about. So, if your even money virtual horse fails to win, it’s not like the rest of the field are big outsiders and he’s been pipped by a 5/4 shot – chances are he’s been beaten by just one of a range of horses in the 4/1 to 14/1 range. Thus there are several more similar horses who could also beat him/her/it.

The problem with your theory here is that first of all, while all results are determined by a random number generator, they don’t follow the same rules as real life. Thus if the even money horse doesn’t win, we have no way of knowing what his actual chances are of coming second. In real life, we know that an even money chance in a twelve furlong high grade flat race is likely to finish, and finish well. Chances of coming fourth or fifth of five runners are slim. Conversely, an even money favourite in a two mile novice chase has a much bigger chance of coming out of the places, due to falling or stumbling at a fence. Without knowing the exact way the algorithms work for the various games that are in play, we can’t answer this definitively, except to say that it would be a surprise if there was sufficient juice in the refund offer to counteract the over-round.

Best stick to the five runner races featuring actual horses!



About Kevin Egan 130 Articles
Kevin Egan is one of Ireland's leading GAA tipsters. He has previously served as GAA odds compiler for a number of major bookmakers.

2 Comments on Every Second Counts

  1. Thanks Kevin!

    With regards to the “2nd – money back” offer on actual horses the favourite can be any price and does not have to be greater than 3/1.

    In the virtuals you are right to say that the odds are skewed so that you dont get any each way value (no odds on fav in a dead eight race here!) but you do get a race similar to the 1.53 in Portman Park with 8-9 “runners”, a afav at 11/8 and 2f at between 3/1 and 9/2.

    I know we cant be sure but I would have thought that the random number generator that would decide on the winner would also decide on the second and the third. So if lets say the 2f won then that is taken out of the book and it gives the 11/8 fav a better than 11/8 chance of finishing 2nd.

    Certainly from what I have seen generally if a short priced fav doesnt finish in first place it seems to finish either 2nd or 3rd an awful lot of the time. I certainly dont think there is anything random in the placings.

    Anyway roll on 1.53!!

    • I will definitely look into the algorithms behind the virtuals and see if anything can be extracted. When it comes to each way betting it’s clearly relevant anyway, so it’s well worthy of study. Best of luck with your punting!

Leave a Reply

Your email address will not be published.