betpilot.club
Newbie
Hi all,
I've been lurking here for a while and wanted to share something I've been
working on, mainly to get some honest criticism from people who actually know
what they're doing.
About two years ago I started training a model to predict football results. It
began as a personal project. I wanted to see whether I could get to something
that holds up long term rather than just having good weeks. It's grown into a
small web app since then, and I've made it free. I'm not selling anything,
there's no paid tier, and I'm not asking anyone to sign up for a tipster
service. I built it for myself and figured others might get something out of it.
What it currently does:
Daily football predictions across the major leagues.
NBA predictions.
Automatically generated parlays and accumulators from the day's picks.
A monthly evaluation page so you can see how the model actually performed,
including the losing months.
That last point is the part I care about most. Everything is timestamped before
kick-off and graded automatically afterwards, so nothing gets quietly deleted
when it goes wrong. I'd rather show a flat month honestly than post screenshots
of winners.
I'm under no illusion that beating the bookmakers long term is easy. That's
exactly why I'm posting here instead of just carrying on alone. What I'd really
value:
Does the way I'm presenting results look credible to you, or does something feel
off?
Am I measuring the right things? I track yield rather than hit rate, since
average odds matter more than how often you're right.
Anything obviously missing that you'd expect from a tool like this?
It's at www.betpilot.club if you want to have a look. Happy to answer anything
about how the model works or what data it uses, and equally happy to hear that
it's rubbish, as long as you tell me why.
Cheers
I've been lurking here for a while and wanted to share something I've been
working on, mainly to get some honest criticism from people who actually know
what they're doing.
About two years ago I started training a model to predict football results. It
began as a personal project. I wanted to see whether I could get to something
that holds up long term rather than just having good weeks. It's grown into a
small web app since then, and I've made it free. I'm not selling anything,
there's no paid tier, and I'm not asking anyone to sign up for a tipster
service. I built it for myself and figured others might get something out of it.
What it currently does:
Daily football predictions across the major leagues.
NBA predictions.
Automatically generated parlays and accumulators from the day's picks.
A monthly evaluation page so you can see how the model actually performed,
including the losing months.
That last point is the part I care about most. Everything is timestamped before
kick-off and graded automatically afterwards, so nothing gets quietly deleted
when it goes wrong. I'd rather show a flat month honestly than post screenshots
of winners.
I'm under no illusion that beating the bookmakers long term is easy. That's
exactly why I'm posting here instead of just carrying on alone. What I'd really
value:
Does the way I'm presenting results look credible to you, or does something feel
off?
Am I measuring the right things? I track yield rather than hit rate, since
average odds matter more than how often you're right.
Anything obviously missing that you'd expect from a tool like this?
It's at www.betpilot.club if you want to have a look. Happy to answer anything
about how the model works or what data it uses, and equally happy to hear that
it's rubbish, as long as you tell me why.
Cheers