The roadmap
Hello everyone,
I'd like to share with you how I envision the upcoming improvements to the site.
Right now, what's mainly coded is nicely displayed statistics, though there's plenty of room to improve them, particularly to make the visual readability easier. This gives players access to new data that doesn't exist on other sites, but it's still up to them to do the analytical work, even if the scores already go a long way toward guiding that step.
The current step is therefore to finish the 'statistics' panel and the other categories that are currently displayed as a simple table. I'd like to take the time to render them as more visual panels, like 'impact' or 'ganks'. The goal is not to be reading an Excel spreadsheet (well... maybe a little). Each 'category' takes roughly 1 to 3 weeks to build, depending on what I add along the way and the difficulties I run into.
After that, I'll have a choice to make: either I finish the 'timeline', or I work on the 'synthesis'.
What is the timeline? It's a 'page' that presents a Riot-style timeline where you can click on a key event: a death, a drake kill, a teamfight, a decisive macro moment. A text is generated so the player can understand the stakes, the risks, and why or how that was a debatable choice (or, on the contrary, a good one). Long generated texts are complex because they require weaving together a lot of variables, and the more variables you add, the more sentence fragments you multiply and the more combinations you have to handle.
Alternatively, I could work on the 'synthesis' (which is my personal favorite). It would be a third analysis 'page' (alongside 'statistics' and 'timeline') containing... a synthesis: the player's strengths and weaknesses, natural-language explanations of the flaws, and suggested exercises to improve in future games, that kind of thing. It feels more accessible and, well, more entertaining to code for me. It'll probably take about 3 months of work, but that's... my problem, and my pleasure :)
Honestly, what I'd really love to tackle after all that is probably the multi-game profile. We aggregate the games you submit for analysis and shine a light on your strengths, your weaknesses, recurring mistakes, and so on. For example, I've analyzed some of my own games alongside those of Agurin (a top jungle challenger), and the differences are OBVIOUS just from looking at the stats: I gank more often, often better, sometimes worse, but he has a monstrous consistency that makes him extremely reliable from one game to the next. I know what I need to work on: making my jungle clears more consistent, stopping myself from forcing the first drake, and ganking a little less. I need to prioritize respawning my jungle camps rather than giving in to my urge to chase kills.
Also, a glossary is in progress, but I haven't really had time to write the cards inside it yet. Woopsie.


