Hi folks, I've been a fan of Indie Hackers since its inception and met a few of you recently at the Dublin meetup / Stripe talk, which was excellent.
Hi folks, I've been a fan of Indie Hackers since its inception and met a few of you recently at the Dublin meetup / Stripe talk, which was excellent.
Since then I've been working on my project, RossTennis.com which is now finally live for v.1.0.
If you are interested in the background... I built a predictive tennis betting model over around a year and a half ago and it's now winning at approximately 12.5% flat rate per bet - the secret was figuring out which matches to bet on and which to avoid. So the model went through many iterations to be able to get to where it is now.
I then teamed up with an ex-colleague of mine to build a front end. Previously the whole thing was running on google sheets. I also brought on around $25k investment and integrated with Sport Radar so that our custom stats will update automatically.
If you are into pro tennis it's pretty fun to play around with the matchups, you can even see how a male player might perform against a female player.
On the business side of things we will make money from:
- affiliate revenue
- b2b data licensing
- our betting fund
- maybe selling tips, to be decided.
It would be awesome if you'd like to check out the site and offer any feedback. We're trying to improve it on a daily basis.
If you like to bet, I recommend a bet on Pella to beat Lorenzi tonight, and Mayer to beat Delbonis tomorrow.
In the mean time, keep on shipping!
Ross.
How easily can you port this to another sport? Cricket betting has been generating some sweet traffic.
I think I could implement a new sport about every 30 days. Cricket is def in my top 5.
Hey interesting product. Gambling is not something I'm personally into but I know Ireland has a big market for this - although not sure how big the tennis betting scene is; I usually only see lads betting on football or horses really. But don't take my word for it as I'm not in that scene anyhow.
Quick question, as a fellow Eire head; you mentioned you brought on some funding - mind me asking how you did that?
I had a startup a number of years ago and we ended up going for the CSF (€50k) and got it - but it was a complete disaster honestly. Would not recommend. It was a duo team and I was the solo dev with no clear product roadmap and we were just scrambling for any bit of traction. No product + no traction is just insanity. How we even got the CSF... I'll never know.
Would love to hear more - keep in touch!
Edit: Also, your Pella tip was right lol.... just looked it up!
Hi Shane, thanks for your feedback. Regarding the funding it was all from one old friend / colleague. I checked out a lot of the public funding stuff but I wouldn't have qualified for their offers. I also tried one or two Irish VC's but to no avail.
Survais looks cool too, nice work, v useful I'd say.
Big congrats landing that funding! Yeah I did meet some VCs a few years back for an old startup - it wasn't a great experience either, although I can't really blame them either.
I'm actually looking for early test users for Survais. If its something you might use let me know and I'll hook you up with extended trial access (full features etc of course) =)
Thanks a lot, great, I'll definitely look at it again once I have a few users to test it on.
Also, I took the bet on Meyer off, it's not as good as I thought it was previously, so better to give it a miss.
Awesome, cheers!
Also just want to pick your brain about your algorithm(s) and if you have plans to expand. Would you be able to enable new avenues or sports for this? Or are there very specific things in Tennis that you are looking at to calculate players incentive to win?
I imagine other two player sports could be a way to expand (such as boxing, mma, darts, squash etc.) - or what about something like Golf, which is understandably different than the others - but interesting to think about nonetheless.
Yeah aiming to expand to multiple sports later this year, even multi-person, like Rugby.
The algo took a long time to get right, and analyses around 75 data points per player.
Very cool, will definitely be keeping an eye!