Exactly my point, the last price traded means nothing in the scheme of things. It is not indicative of any trend, it does not point to where the bulk of the action happened, and it can be so far out that it's not funny.

On numerous occassions I have noted that the bulk of activity might have been at the $3.00 amount for tens of thousands of dollars, the a $24.00 bet (The last bet recorded) might have been at $4.50 because someone was just plain desperate with only a split second to go before the jump & Cancel sequence.

This would then by their reckoning mean the 'Starting price' was $4.50 when in fact it should have been $3.00. This is a flawed model they are using.

I have recorded some final trade prices on Betfair for the selections in my Lay systems, and in every instance the price shown on Betfair Advantage is lower than my recorded price. I can't remember which particular selection it was but the final Betfair trade price was slightly inxs of $3.00 yet the price shown on the B.A. site was $2.22. In more instances than not, the difference is not all that great but as I said it has always been lower.

Maybe the real final trade price is known only after the records on the Betfair page for a particular race are no longer accessible to the public?

I sent their customer service an e-mail a few days ago about this perceived anomaly and they replied it will be referred to their technical people.

Apropos the above posts the final trade price is wrong, often way out, and anyone who thinks that Betfair will make any effort to fix it obviously is unaware of their abysmal track record. Betfair fixes stuff that isn't broke, thus breaking it, and then doesn't bother fixing the broken result.

Furthermore that final and current trade price is typically useless.

The correct figure that should be displayed alongside the current price is the weighted harmonic mean of matched bets.

This is easy to do, and is essentially what Betfair was proposing as their replacement for the offical SP.

However they show every sign of avoiding that obvious necessity. Presumably they are afraid that instantly being able to assess the true price will help inexperienced players, thus making it harder for the experienced well-informed sharks to rip the newcomers off.

My tip to newcomers is not to get sucked in by what appears to the current price but instead set your own price at what appears to be favourable to you. This means that sometimes your prices won't get matched, but the better prices you do sometimes get more than makes up for not getting set.

I have received a reply from Betfair.

The price shown on Betfair Advantage is the final Back price displayed on the screen at the time that betting on a race is suspended.

I looked at an example where the final trade price on a particular horse was $100 but because everything available at this price was snapped-up at the death, the next available Back price was $85 - it is this price of $85 that is shown on Betfair Advantage. As a matter of interest, the next available Lay price at the time of suspension was $160.

I assume this is the reason why the final trade price can often be higher than that shown on Betfair Advantage.

TailWag,

Anyone uncertain what a weighted harmonic mean is can easily look it up. Whereupon they will note that your formulas are not weighted.

The well-known fact that Betfair choose to inflate its matched bets figure by doubling them, will not alter the end result - as you yourself noted.

-----

Meanwhile anyone troubling to study the above might find this paradox about weighting cute:

http://en.wikipedia.org/wiki/Simpson%27s_paradox

Yes you are correct, these are 'Simple' formulas, which implies that one would know the difference between 'simple' and 'complex'. Anyhoooo the essential point is that Betfair has a flawed premise which will floor you over time. (pun fully intended).

Woof
Tailwag

In last night's 7 race program at Canterbury 3 winners paid an unflattering $1.04 for the place.

By my figures such a record-equalling farcical situation occurs less than once a year on average. Ominously 2006 managed that feat 5 times.

But the BAT for the relevant races 2,3 & 7 shows divs of $1.16, $1.18 & $1.13.

A staggering 292% extra profit.

Tempting as it is to dream up some sure-fire system from this, my only safe conclusion is that those $1.04 backers were not very bright.

Hi, jfc.

It's not just last night that this "extra profit" occurs.

Every instance I've looked at I've noticed that the smaller the price on the TAB then the better it is percentagewise on Betfair. It must be a brave person who is offering such a high Lay price hoping that it is matched.

I just had a thought (interesting and dangerous) about what would be useful information from Betfair. Working on the averages model rather than the last price taken, it would be useful to know the number of unique bets i.e. that is the number of people who bet, not the total number of bets.

A large punter (in terms of volume) with the wrong opinion in a race can cause the sentiment in the race to be significantly altered. However by knowing the total number of unique punters who are laying as opposed to backing, you can then work out a true weighted average by 'Interest' as opposed to a straighout numeric figure.

In reality what matters is not the size of a persons bet, but whether or not they are correct (win or lose), so knowing the true level of that sentiment regarding any individual horse is paramount, but lost to us because that information is not displayed, but it is obviously known.

Do you think Betfair would ever give that sort of 'Advantage' to us - normal punters? it would make their tool very usable and significantly better than their flawed current model.

Comments?

Tailwag.