Spitballing idea to collect handicapping data in a crowd sourced way

[JerryBoyle]

The recent thread on sizing trifectas got me to thinking about building a small calculator for sizing bets using various sizing algorithms, which lead me to thinking about crowd sourcing some handicapping data. My thinking is that it might have 2 advantages:

1. The data can then be published for free without Equibase complaining
2. The data can be "moved" towards being correct (and correct over time)

First, a simple example using track take out rates. For anyone who uses these rates in any systemized way, it can be a pain to initially source them all, and it's nearly impossible to keep them up to date as tracks change them. Many of the above mentioned bet sizers use the takeout, so the calculator could include something like a track code drop down, which underneath the hood of the app links to the takeout rate stored in the application. If the user disagrees with the takeout rate that is presented for the selected track, either because it's wrong, or because it was recently changed by the track, or because it hasn't been sourced yet, the user can input the new rate. Behind the scenes, whichever rate the user enters acts like a vote for that rate. The app can then use that info to determine the appropriate rate.

The above example primarily illustrates how the 2nd bullet might be accomplished and, for that specific case, has the benefit that it's generated off of users using another feature - the bet sizer. That is, it doesn't require good citizens to feel like going to some website to update the rate changes. I'm struggling to think of how we might apply the same idea to data that is published by Equibase. Could we do the same with POC times somehow? Maybe good citizens go to a site to input the "correct" times or lengths ahead/behind, that information then becomes published for free consumption? Like the takeout example, the inputs act like votes which push towards the correct answer? This could be done for distances for tracks as well. E.g. an "about 7f distance" is voted on to be 1500yds at Gulfstream. The issue I'm struggling with is how do we get the initial publication or vote up such that every single race isn't missing? Obviously, the starting point might be the Equibase #'s, but this is obviously frowned upon by Equibase.

What say you? Worth some more discussion, totally worthless, unintelligible?

[Suff]

I looked, could'nt find the trifecta sizing thread. You're suggesting crowd sourced inputs into an application? Jerry Brown from Thorograph was on @athepostlive podcast this week.

He said he has 10 fulltime people, and 40 part-time at Race Tracks. The PT employees at the tracks input granular data. Example, every race's wind speed and direction is recorded. You're idea is doing that at scale using people, who'd do it free? Because of the direct benefits or as a rabid enthusiast?


This is totally off subject, except to speculate that the Algorithms to operate Single Pool Wagering , could be repurposed to perform your tasks.



Single-pool wagering could conceptually allow for the creation of new bet types, since the agglomeration of all bets in one pool would allow tracks to offer wagers that would not normally draw enough money to ensure a segregated pool’s liquidity – in other words, providing protection to bettors who do not want large wagers to dramatically impact their payoffs. For that reason, single-pool wagering is favored by teams that operate robotic wagering programs that dump hundreds of bets into the pools at the last moment before the pools are closed. The extraordinary access of the robotic programs to the racing industry’s pool data and the use of lucrative rebates to keep the programs in business is a controversial topic in racing, especially among the sport’s most critical customers.