Year In DotA – 2014

I am a support player at heart. Not for any altruistic or sympathetic reasons–not because I have a need to help my teammates, or a secret fear of playing carries. I am a support player at heart simply because there are maybe 3 carries in the game that I enjoy playing, and even fewer mid or offlane heroes, while I can find merit in most every hero that is traditionally put in a support role, and so a strong desire to play them.

Further, it was my desire to prove that I could support my way up the MMR ladder, where many others thought the only way to climb was to pick core heroes and carry games. I mean, it’s not as if I didn’t do that myself. Even now, after having barely player her for a year, Queen of Pain is my most played hero (590 games, with the next highest at only 411). All the pubstomping I did with her gave me, even after a monumentally bad losing streak, a comfortable 4600 MMR rating when I first entered ranked play.

Anyway, let’s look at a graph.

graph

What interested me most when I first saw this graph was that I could immediately match the biggest peaks and plateaus with real life events. For example, the period from mid July into mid August is when my PC first hacked up a lung and died on me. After having that fixed, there’s that abysmal fall in August, when I was within 3 wins of 5k. That’s when I got very, very sick, and decided to play anyway. It did not go well, as is obvious now, and I only stopped hemorrhaging MMR points when my PC died completely, forcing me to stop playing. That’s also when 6.83 was released, and I was sidelined for over a month while players became acclimated to one of the biggest patches DotA has seen in years.

There is a noticeable difference in the quality of the games I ended up in once I fell below about 4.5k. Above that it was common to get games with many 5k players, and even 6k players, depending on the hour. Supporting in those games was a breeze. I could depend on a carry that I secured farm for to get good items at a decent pace. I could rely on the mid player to at least hold their own in lane, and for the offlane players to not feed. Everyone did their jobs, and I knew that if I played well it would have a positive and measurable impact on my team’s ability to come out ahead. Isn’t that what everyone wants? I mean, it’s not as if I didn’t lose games where I played well, or won games where I fed like an idiot, but for the most part I knew that if I did what I was there to do, I could rely on everyone else to at least attempt to do the same.

When I fell below my starting MMR, and even lower than that, the games started to get bad. I won lanes for carries who couldn’t farm, or who bought the most boneheaded items possible. A Luna who went Midas into Shadow Blade and proceeded to run away from every fight. A Phantom Assassin who rushed a Diffusal Blade and then did nothing. The MMR distributions in those games is a lot safer, and trends downward instead of up. I started to be matched  with 3k players, and often I found at the end of a losing game that I was the highest MMR player on my team, which is off-putting when I’m also the only support.

My ranked win rate dropped below 50% for the first time, which was discouraging alongside the need to catch up with other players when I put my PC back together and entered 6.83 a month late. But I struggled on anyway, and even went back to playing support. I am not one to shrink away from something as petty as losing. I know that to ultimately succeed I have to play better than I was, and there’s every chance that would mean throwing away several hundred more MMR. Losing is something I’m used to, and I get over it easily.

winrate 1

There’s one more oddity.  Over the last 12 months I’ve also had 60 more Dire games than Radiant. During that same period I have a ~49% win rate in ranked matchmaking across ~670 games. Which is fine. I expected as much. However, I also have a a ~52% win rate on Radiant, compared to a ~46% win rate on Dire. This is in line with my historical average of roughly 48% win rate on Dire and 53.5% win rate on Radiant. This month alone I am under 40% on Dire and nearly 60% on Radiant. What gives?

winrate 2

It only makes me wonder how much the matchmaking system takes that sort of thing into account. Could it be that if I’d been given as many (or more) Radiant games as Dire I’d be well ahead? Might those 60 games have put me above 5k by now? Probably not, at least not completely. I wasn’t cheated out of anything. I played poorly, for whatever reason, and I lost a lot of games. Still, it sure would be nice to get those extra Radiant games.

The final question raised is, of course, why? I cannot say. There is nothing in my play that stands out as being more acclimated to playing on Radiant over Dire, except that Radiant itself has almost always had a higher win rate. Looking over the stats of some players I’ve been matched with, most of them are also winning more on Radiant, but only one other had such a drastic discrepancy between how often they get put on each side and how often they win or lose, and that player was 41% on Dire and 50% on Radiant. Like nearly everyone else, they also had more games played on their losing side. In fact, my informal survey showed the majority of players having more Dire games, at a lower win rate, with only a few players being lucky enough to get a lopsided amount of Radiant games. For my part, I tend toward believing that I am just terrible on Dire rather than being amazing on Radiant. Clearly the answer is not to hope I win the Radiant lottery, but to instead get better at playing on Dire. Perhaps easier said than done, since I don’t yet know what I’m missing. At least I have something to look out for, an aspect of my game to improve beyond generally not being so bad.

Anyway, 5k support is still the goal, and once I’ve managed to pull myself back up to reasonable games, I firmly believe I will get there.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s