Follow-Up Discussion on Analytical Content

After posting my video the other day about different analytical content templates and “win conditions analysis,” I had a conversation with Paolo “Sandata” Bogo, an esports journalist from the Philippines. Paolo had some good follow-up questions, and he was willing to let me share my answers here in case anyone else was curious about the same things. Thanks for the conversation, Paolo!


I really like the idea of looking at win conditions when making analysis. My question though is: how do you determine from all the data points in MOBAs which ones are really contributing to wins?

Good question. I wasn’t referring specifically to stats analysis. You can use stats to support win conditions, but I think you need to do other kinds of analysis to really get the full picture.

Statistically you might be able to say “This team wins 90% of the time when they get first blood,” but you have to look beyond that and try to figure out why that’s the case. Are they good at being aggressive when they have the lead? Do they know how to siege well, or do they tower dive a lot and their tower dives work a lot better when they have an advantage already?

You can’t really learn some of those things with stats alone. You can use data to find some points of interest, but the data won’t give you the full story. So it depends on how you want to communicate the win conditions. If you want a few quick sound bites, you can do stats-based win conditions. If you want a full article, you need to combine the numbers with other types of observations.

Fair enough. I think where I’m getting tripped up is on the execution. Are you saying its a case of observing first how a team wins, then going over to the stats to verify that? I guess I’m asking because I’m trying to incorporate more analysis into the work that I do, but I always get that nagging feeling that maybe I’m letting my own biases into the analysis, diluting it somewhat from something actually valuable or true.

Sometimes it starts with seeing something in the numbers, then going back to the VODs or your notes to investigate. Sometimes you’ll see something while watching, and that makes you ask a question that you go and check out in the stats. It can go both ways.

Analysis can be nerve-wracking. It is for me. You will always naturally watch some parts of the game more than others. Ultimately you have to get over the fear of being wrong or biased, and just try to be confident and support your arguments with evidence. Learn from your mistakes; don’t take criticism personally.

What do you think about analysts and analytical content that ends with the author [making a prediction]? It’s a bit of polarizing topic for me.

I think predictions are a good way to end some types of articles. It depends on the article. If you are going to make a prediction, I think you need to write the article with that in mind. The reader should have an idea that the prediction is coming, so that it doesn’t seem out of place.

Predictions can pair well with win conditions. You go through some win conditions for each team, then conclude by making a prediction that is based on how much confidence you have in each team to meet their conditions. Or you can make the prediction the core of the piece. State your prediction up front, then spend the rest of the article explaining why you think that way. Give reasons why that team or player will win, and/or reasons why the other team or player will lose. You can argue against yourself a bit, too: “I’ll be wrong if…” That is kind of like looking at win conditions, if you go in that direction.

Magic on EmailMagic on FacebookMagic on TwitterMagic on Youtube
Tim “Magic” Sevenhuysen runs, the premier source for League of Legends esports statistics, and writes for theScore esports.

Leave a Comment