@mersecske

#DEVLOG 4#
Okay, guys, I’ve finally found the cause of the issue.
It was not in the code, it was in my mind.🤦‍♂️If I wanted the model to predict a draw, it would lose by accuracy.
To understand this I made some calculations:

away: 7341  matches  - 31,6%
draw: 5522  matches  - 23,8%
home: 10368 matches  - 44,6%

SUM: 23231 matches

In a nutshell: It NEVER worth it to predict a draw!

Now that I “solved” this problem, I can focus on increasing accuracy: adding more aspects, finding bugs, etc.

#DEVLOG 4#
Okay, guys, I’ve finally found the cause of the issue.
It was not in the code, it was in my mind.🤦‍♂️If I wanted the model to predict a draw, it would lose by accuracy.
To understand this I made some calculations:

away: 7341  matches  - 31,6%
draw: 5522  matches  - 23,8%
home: 10368 matches  - 44,6%

SUM: 23231 matches

In a nutshell: It NEVER worth it to predict a draw!

Now that I “solved” this problem, I can focus on increasing accuracy: adding more aspects, finding bugs, etc.

#DEVLOG 3#
This is just a quick devlog from me.
I have some issues. The main thing is what I have already written about: the model NEVER EVER predicts a draw. (It is because of the mechanism of Poisson-model…). I tried different methods to solve this problem, but none of them worked.
There is 2 ways to solve this:

1. Get a bigger dataset. But the API that I’m using provides only 100 calls/day which is a really small number. So I would need to pay, which I don’t really want. (In the very unexpected case you want to fund my project, DM me @be.the.nike)
2. There is a model that corrigates the torsion of Poisson-model. BUT AI said to me that to achive this I will have to kind of write my own AI model training algorithm. And this sounds really scary to me

Cheer for me!

Some good news at the end: my model is a tiny little bit better in predicting the winner of a match than betting offices!

#DEVLOG 3#
This is just a quick devlog from me.
I have some issues. The main thing is what I have already written about: the model NEVER EVER predicts a draw. (It is because of the mechanism of Poisson-model…). I tried different methods to solve this problem, but none of them worked.
There is 2 ways to solve this:

1. Get a bigger dataset. But the API that I’m using provides only 100 calls/day which is a really small number. So I would need to pay, which I don’t really want. (In the very unexpected case you want to fund my project, DM me @be.the.nike)
2. There is a model that corrigates the torsion of Poisson-model. BUT AI said to me that to achive this I will have to kind of write my own AI model training algorithm. And this sounds really scary to me

Cheer for me!

Some good news at the end: my model is a tiny little bit better in predicting the winner of a match than betting offices!

#DEVLOG 1#

I started to work on a website that predicts the scores of a football match. I’ll make my own AI model to create it.
To do so my first task is to get a database. There are a lot of APIs on the internet but just a few of them provides a service without any limitations. I’ve finally find one on Github. So I started to work on it and format the data. BUT after ~2 hours I realized, it wouldn’t work.
The reason why it wouldn’t work: I wrote some criteria that this AI model should consider. And the Github repo did not fit to it.
Sooo I’ll start over with a new provider.
By the way, here are the aspects I was talking about:

#DEVLOG 2#

I haven’t posted devlog for a while. But I was working hard.
I ran the secind evaluation of the Poisson regression model (the first model was so unreliable that it don’t desire to mention).

Results:

• Outcome (H/D/A) accuracy: 54.1% ± 2.1%
• Goal-count prediction (Poisson deviance): 1.13
• Feature importances look sensible: team_attack, opp_defense, is_home are the strongest signals
• Weakness: model almost never predicts draws I tried a “draw margin” trick, but it didn’t work

Next options (ideas only, none started yet):

• Add Elo rating (instead of current attack/defense strength method)
• Dixon-Coles correction to address draw underestimation
• Incorporate missing key players as a feature (it costs too many API calls)
• Tune alpha hyperparameter

This is my first project written in Python so I’m still learning pretty much. I’ll push my progress to Github as soon as I can produce accurate enough predictions. Be patient, I’m doing my best!

Arrivederci, Auf Wiedersehen, Goodbye, A viszontlátásra!!

#DEVLOG 2#

I haven’t posted devlog for a while. But I was working hard.
I ran the secind evaluation of the Poisson regression model (the first model was so unreliable that it don’t desire to mention).

Results:

• Outcome (H/D/A) accuracy: 54.1% ± 2.1%
• Goal-count prediction (Poisson deviance): 1.13
• Feature importances look sensible: team_attack, opp_defense, is_home are the strongest signals
• Weakness: model almost never predicts draws I tried a “draw margin” trick, but it didn’t work

Next options (ideas only, none started yet):

• Add Elo rating (instead of current attack/defense strength method)
• Dixon-Coles correction to address draw underestimation
• Incorporate missing key players as a feature (it costs too many API calls)
• Tune alpha hyperparameter

This is my first project written in Python so I’m still learning pretty much. I’ll push my progress to Github as soon as I can produce accurate enough predictions. Be patient, I’m doing my best!

Arrivederci, Auf Wiedersehen, Goodbye, A viszontlátásra!!

#DEVLOG 1#

I started to work on a website that predicts the scores of a football match. I’ll make my own AI model to create it.
To do so my first task is to get a database. There are a lot of APIs on the internet but just a few of them provides a service without any limitations. I’ve finally find one on Github. So I started to work on it and format the data. BUT after ~2 hours I realized, it wouldn’t work.
The reason why it wouldn’t work: I wrote some criteria that this AI model should consider. And the Github repo did not fit to it.
Sooo I’ll start over with a new provider.
By the way, here are the aspects I was talking about:

Today I got a feedback, that said, I should translate my site so people on Hackclub could understand, what is written here.
If I had choosen the elegant way to do this, it would’ve taken a plenty of time. And for me it would’ve been wasted time, because this site was developed for Hungarians.
So I just added a quick Google Translate widget. It is pretty ugly, but you can understand it now.
I hope you like it!
In case you found a bug, please DM me on Slack: @be.the.nike

Today I got a feedback, that said, I should translate my site so people on Hackclub could understand, what is written here.
If I had choosen the elegant way to do this, it would’ve taken a plenty of time. And for me it would’ve been wasted time, because this site was developed for Hungarians.
So I just added a quick Google Translate widget. It is pretty ugly, but you can understand it now.
I hope you like it!
In case you found a bug, please DM me on Slack: @be.the.nike

Devlog — Sportválasztó (July 2026)
The last two days were all about squashing login and connectivity bugs on sportvalaszto.hu. Here’s what went wrong and how I fixed it.
Bug #1: The site wouldn’t load for some international testers
Turned out appwrite.js had a hardcoded endpoint (cloud.appwrite.io/v1) that didn’t match the region-specific one in .env (fra.cloud.appwrite.io/v1). This mismatch broke the app for some foreign (Hack Club) testers.
Fix:

Swapped the hardcoded value for import.meta.env in appwrite.js
Made sure the hosting provider had the VITE_APPWRITE_* env vars set at build time, not just runtime

Bug #2: Google OAuth redirect_uri_mismatch
Login was failing with a redirect_uri_mismatch error from Google.
Fix:

Registered the full callback URL in Google Cloud Console under Authorized redirect URIs:
https://fra.cloud.appwrite.io/v1/account/sessions/oauth2/callback/google/68fe2fae00030619f0a5
Kept Authorized JavaScript origins domain-only (no path):
https://fra.cloud.appwrite.io

Mixing these two up was the root cause — an easy trap to fall into.

Stack: Appwrite Cloud (Frankfurt region) · Google OAuth 2.0
Both bugs are now resolved and login works reliably for testers regardless of location. Hurrayy!!

Devlog — Sportválasztó (July 2026)
The last two days were all about squashing login and connectivity bugs on sportvalaszto.hu. Here’s what went wrong and how I fixed it.
Bug #1: The site wouldn’t load for some international testers
Turned out appwrite.js had a hardcoded endpoint (cloud.appwrite.io/v1) that didn’t match the region-specific one in .env (fra.cloud.appwrite.io/v1). This mismatch broke the app for some foreign (Hack Club) testers.
Fix:

Swapped the hardcoded value for import.meta.env in appwrite.js
Made sure the hosting provider had the VITE_APPWRITE_* env vars set at build time, not just runtime

Bug #2: Google OAuth redirect_uri_mismatch
Login was failing with a redirect_uri_mismatch error from Google.
Fix:

Registered the full callback URL in Google Cloud Console under Authorized redirect URIs:
https://fra.cloud.appwrite.io/v1/account/sessions/oauth2/callback/google/68fe2fae00030619f0a5
Kept Authorized JavaScript origins domain-only (no path):
https://fra.cloud.appwrite.io

Mixing these two up was the root cause — an easy trap to fall into.

Stack: Appwrite Cloud (Frankfurt region) · Google OAuth 2.0
Both bugs are now resolved and login works reliably for testers regardless of location. Hurrayy!!

Devlog: Sportválasztó
What is this project?
Sportválasztó (Hungarian for “Sport Selector”) is a website I’ve been building to solve a problem that’s been bothering me for a long time: in Hungary, finding a sports club is genuinely awful. There’s no good, centralized, user-friendly way to browse sports and find a club that actually fits what you’re looking for. So I decided to build one myself.
Timeline
I actually started working on this project last fall, long before I joined Hack Club. It’s been a side project I come back to whenever I have the time and motivation for it, so progress has been steady but not constant — more of a “work on it in bursts” kind of project than a daily grind. Joining the Stardance Challenge felt like a great excuse to pick it back up with more focus and actually push it forward.
Tech stack
I’m keeping things simple and dependency-free:

HTML for structure
CSS for styling
JavaScript — all vanilla, no frameworks or libraries

No React, no build tools, no bloat. Just the raw basics, which honestly was not the best decision, but at the time I started I had no clue about any frameworks existing.
A small honest note
One thing worth mentioning: the site itself is in Hungarian, since it’s built for Hungarian users trying to find local sports clubs. I know Hack Club’s community and platform are primarily English-speaking, so there’s a bit of a language mismatch between the project and the challenge. I’m keeping the devlogs in English, but the actual product stays Hungarian, because that’s who it’s meant to serve.
What’s next
I’ll keep posting updates here as I keep building — adding features, improving the design, and (hopefully) making it easier than ever for people in Hungary to find a sport and a club that’s right for them.

Devlog: Sportválasztó
What is this project?
Sportválasztó (Hungarian for “Sport Selector”) is a website I’ve been building to solve a problem that’s been bothering me for a long time: in Hungary, finding a sports club is genuinely awful. There’s no good, centralized, user-friendly way to browse sports and find a club that actually fits what you’re looking for. So I decided to build one myself.
Timeline
I actually started working on this project last fall, long before I joined Hack Club. It’s been a side project I come back to whenever I have the time and motivation for it, so progress has been steady but not constant — more of a “work on it in bursts” kind of project than a daily grind. Joining the Stardance Challenge felt like a great excuse to pick it back up with more focus and actually push it forward.
Tech stack
I’m keeping things simple and dependency-free:

HTML for structure
CSS for styling
JavaScript — all vanilla, no frameworks or libraries

No React, no build tools, no bloat. Just the raw basics, which honestly was not the best decision, but at the time I started I had no clue about any frameworks existing.
A small honest note
One thing worth mentioning: the site itself is in Hungarian, since it’s built for Hungarian users trying to find local sports clubs. I know Hack Club’s community and platform are primarily English-speaking, so there’s a bit of a language mismatch between the project and the challenge. I’m keeping the devlogs in English, but the actual product stays Hungarian, because that’s who it’s meant to serve.
What’s next
I’ll keep posting updates here as I keep building — adding features, improving the design, and (hopefully) making it easier than ever for people in Hungary to find a sport and a club that’s right for them.