Front Runners

[MrJoeBlack]

I’ve been running an automation for about six months with good results on the flat, but I’m running into a consistent issue on the jumps. Front-runner sentiment seems to create big distortions late in the pre-race markets.

My historical data seems to indicate that there is a bot that kicks in 2 mins before the off - however the market seems to adjust to this relatively quickly. However, I am yet to find a way to account for the 'front runner money'. It’s easy to handle manually because I can work off of visual clues but I would need timing data (which is rarely consistent) for my automation.

Has anyone else noticed this timing pattern or found ways to adapt their automation around it? Would love to hear how others tackle the “front-runner money” problem.

[jamesedwards]

I’ve been running an automation for about six months with good results on the flat, but I’m running into a consistent issue on the jumps. Front-runner sentiment seems to create big distortions late in the pre-race markets.

My historical data seems to indicate that there is a bot that kicks in 2 mins before the off - however the market seems to adjust to this relatively quickly. However, I am yet to find a way to account for the 'front runner money'. It’s easy to handle manually because I can work off of visual clues but I would need timing data (which is rarely consistent) for my automation.

Has anyone else noticed this timing pattern or found ways to adapt their automation around it? Would love to hear how others tackle the “front-runner money” problem.

I'm guessing the effect you are seeing is people backing likely front-runners based on how they line up just before the off. Presumably this can only be done manually.

[MrJoeBlack]

Yes, that's exactly it - currently eroding the profit! I'll have to stick with manual adjustments for now.

[jamesedwards]

Yes, that's exactly it - currently eroding the profit! I'll have to stick with manual adjustments for now.

You could potentially do something using GPS tracking through TPD as they line up. Difficult to tell when they are lining up using this method though.