Hi Piers,
Thanks for the question - it's a very good one! We are busy compiling multiple statistics to illustrate the use of the oddsline, and this answer will be updated when there are links to this content. We will write a blog post soon on general advantages, but in the meantime it's worth emphasising that one of the standout features of a machine learning-driven oddsline, on top of being able to spot value bets and being a good indication of any horse's chances, is consistency and predictability.
The Betwise Daily Oddsline, for example, uses models subject to vigorous backtesting so that we can be sure of the edge and strike rates in different probability ranges. Of course that is not to say that using the oddsline produces an automatic profit in certain odds ranges (though some race types and probability ranges do show this), but rather that confidence in strike rates (for example) at certain ranks and probability ranges means that these selections can be further optimised by using the Smartform database and additional data driven rules.
Since we don't use any market data in the oddsline modelling process, it's also a useful take on market movements versus form "picks", as well as a rank ordering tool for exotic plays such as the placepot and trifecta etc.
Hope this makes sense, any other questions please let us know - the new content with statistics to support the above should be available by the end of the month.
Regards,
Colin
