Devlog #7

Sleeper API integration is finally here! This took a while to get out, but I’m so happy that it’s finally here! The main limitation is that it’s based on draft ID, so it might be a hassle to find. However, I can’t control that as Sleeper is read-only API, which means I can’t allow for people to sign in and then access leagues and mock drafts that way. Regardless, it works really great based on my tests with the caveat that if you update the draft settings, you may need to refresh once or twice to ensure they’re updated on ffdrafter. Once you’re in the draft, the main thing you do is simply check back in and view suggestions on the side panel, manually logging picks is disabled since you’re doing that on Sleeper anyway! Another unfortunate caveat is that syncs do take a little bit(5-7 seconds) when others are auto-drafting and when the draft first starts. These are unfortunate side effects, but again they aren’t in my control as I don’t know how often Sleeper updates and sends drafts responses back. Yet, the feature still works even if it does take a small amount of time, so hope it works for anyone who uses it. Attaches is the UI for how the user inputs the Sleeper Draft ID.

Devlog #2

So instead of generating a baseline based off of just personal bio-data, the user goes through a series of drills to see where they stand at currently. They have to practice threes, mid-rangers, layups, and practice some ball-handling drills. Based off of these beginning stats, it generates their current overall and saves their stats to be used later. Right now this is the only main thing that works and based off of this I’m gonna generate a training regime that best fits them mainly off of these stats. Other than that, the user is also able to input their shot diet on the training screen which basically tells the computer what they do when they have the ball. For example, they shoot threes 15% of the time, mid-rangers 35%, layups 40%, and handle the ball via dribbling 10%. Attached is a sample results page after the user finishes their baseline.

Devlog #6

A few more style updates have occurred since my last log. First, the font has changes to be a little more fancy, if that’s what you want to call it, but I think it looks nice for the site itself. Along with that depending on if you’re searching, seeing suggestions, or viewing top talent, different info appears when you click on the player. The commonality is showing the team logo and their tier. But when searching it only shows the top reason, when seeing suggestions it gives age and top 2 reasons, and viewing top talent also shows age along with position(jersey number). Other than that it’s still basically the same since functionality wise, nothing really changed. I hope to start implement the direct Sleeper API integration along with other sites if possible including ESPN and Yahoo Fantasy.

Devlog #5

So I was recently about to ship but it got recalled because I used Claude to do the whole UI and never touched up on it again, which is my fault completely. However, since then I have actually updated the entire UI by adding modular sections and more color by position points. It has the same dark tone as before, but there’s a new purple accent now to make it less generic. Along with that, every position is basically customized by it’s own unique color as well, providing a connection beyond text. Other than that, text was made bigger, and font colors were made lighter to make them pop out more against the dark background. Attached is how the starting page looks like now.

Devlog #4

This devlog is the last before the first ship as the site is now officially live! The only change made between this log and the last is that you can choose whether you have a snake or linear draft. Other than that, hosting is being done through Render(hopefully an accepted hosting platform) and the link is officially live at: https://fantasy-football-drafter.onrender.com/. Attached is how the menu looks now that you can choose draft type.

Devlog #3

So unfortunately, I wasn’t able to connect to Sleeper API, because I was busy updating the value algorithm to make it sure it worked well. Earlier, it was favoring positions too much and wasn’t realizing that starters matter more than bench. I think I’ve found the perfect medium between balancing both maximizing starting talent and a decent bench lineup. I expect to ship within the next few days once I upload a live link and maybe add a feature to differentiate between snake and linear draft along with adding more position customization perhaps. Other than that, this project will probably ship with just a manual mode at first, and then have another ship with Sleeper API connected to it. Attached is a sample of how the UI looks when updating draft picks based on what’s going on in.

Devlog #2

Well, the initial part of the project is completed. There is now a value calculation algorithm in place that ranks each player based on a number of metrics: current draft pick, player ranking, bye week, and roster needs. After that, it gives 8 suggested players in the menu, ranked by points, and also gives warnings about possible reaches and the current pick, but also symbols that could signify a steal at the current pick. Additional testing will be conducted in regards to how well the system seems to be working by simulating a separate mock draft with online resources. My next plan after this, if the system is working correctly, is to connect it seamlessly with Sleeper, as stated in Devlog #1. Attached is how the user interface currently looks after starting the draft.

Devlog #1

This is the start of a fantasy football draft assistant project I’m working on. Currently, it doesn’t do much as most of the time was spent on gathering data for the three main scoring types: PPR, Half-PPR, and Standard, along with the player database from Sleeper. The plan is to eventually find an algorithm to calculate expected value for each player based on different data points: your draft position, current draft pick, player data in general, along with other external factors such as current roster. Also I’m only working on the manual mode right now where all picks have to be manually put in but since Sleeper has a free API(used for player data), I hope to eventually add a seamless integration where it can directly link to a Sleeper League. Provided below is the starter UI menu but will be changed over time as more features are added.

First: Thank you everyone for their reviews on my first ship, this will likely be the last devlog for this project as it is a time-based project(can’t really update after the world cup ends) but I appreciate all the feedback. As for this current devlog, there was a small fix which needed to be done, I originally said on my last post that after a matchup occurs in real-time, it is locked and can’t be simulated anymore. This is not the case anymore and you have the option of locking/unlocking games that have occurred in real time already.

The group stage just completed today, but luckily the knockout functionality was essentially all set to go. After inputting the matchups, the site now has new knockout features equipped. There is an interactive bracket, to simulate your own matchups that includes penalty shootouts, along with a simple page to display advancement probabilities to each round. Another ship should be out shortly to reflect these new changes.

So I’m currently working on a basketball trainer website, eventually converted into an app to help people improve their basketball skills. Currently, there’s only a good UI base because I haven’t started the full functionality yet. I plan on, hopefully, adding a new training section twice every week, or just weekly if that’s what works. Attached is the home page where the user inputs their info to get a baseline for training rigor.

This might be the last devlog for a bit, but the project is officially live with the help of streamlit as this is a python project. The link is already live, so you can simulate your own matches and see what the consensus result was through multiple simulations. This project will be shipped soon but updated again for when the knockout stage matchups are set in stone. Attached is how it looks on the main page.

Update: I added the FIFA rankings as another metric to simulate the games. I also strayed away from using the poisson model to predict outcome and now only use it for scoring purposes as the random forest model is much more accurate. Attached are the updated probabilities.

So I recently decided to train a model to help predict the group stage matches for the FIFA world cup. It’s a combination of a Random Forest Model and Poisson Distribution to help predict the games. Currently I’m only training the model based off of a teams ELO and recent form(performance in past 5 games), but plan on adding FIFA’s point system as another metric. I attached the most common match result of each game after running the simulation 1000 times and the probability of where each team finishes this year(first, second, third qual., third elim., and fourth)