Dota: A Transforming Game


Esports are on the rise, and the fact is beyond doubt. Partly its in the money. In the newly created premier tournaments from The International 5 in Seattle, through the upcoming Manila Major in May, Valve will have payed out twenty-seven million dollars in winnings. The amount of money spent by fans to support the extravagant prize pool, stadiums, and online streaming may still be insignificant compared to the NFL but the rapidness of its increase is astounding (and Valve will never have to pay a settlement for debilitating injury). Additionally, ESPN has started to take notice, including coverage of TI 5, and featuring Dota 2 in their Sports Center Top 10 plays of the night. If you listen to the ESPN announcers in both videos (try to ignore TobiWan in the background), there is a real difference between the two. The woman in the TI5 video may or may not be a fan, but she puts in the effort to understand what's happening and does an exceptional job explaining the plays. On the other hand, the two guys on Sports Center don't understand, don't try, and then blow the whole thing off. Esports fans appreciate professional sportscasters that take the time to understand the sport, and respect it.

What sets Esports aside from regular sports is the constant transformation that the league undergoes. Unlike the NFL, NHL, or other professional sports that occasionally alter the rules, traditional sports can not alter the fabric of the game itself.

Dota has existed since the release of Warcraft 3 by Blizzard in 2002. Warcraft allowed users a generous amount of tools to create incredibly detailed customs games, using the visuals of the base game. Dota was the most popular of myriad custom games. In 2011 Dota finally separated from Warcraft and was remade as a solo game, now called Dota 2. Though it was already successful, Valve continued to release new versions (the first Dota 2 version was 6.70). Since its release, Dota 2 has adopted a new version roughly every four to six months.

The most common change are the nerfs and buffs. Nerfs are improvements to a hero, whereas buffs make a hero less effective. In any patch, at least half of the heroes will receive a minor buff or nerf. How do the developers decide? It seems pretty straightforward, though there is no official policy. Icefrog, the unknown person behind the development, has to consider many angles.

For instance, some heroes are stronger than others. Dotabuff compiles a monthly win rate for all heroes. Seems pretty simple. But it also breaks the win rate down by skill brackets. Looking at this list, it is apparent that a hero's win rate varies by five percentage on average, but by as much as twelve on the extreme cases (currently Medusa and Huskar). Even this is only part of the picture. The professional scene is as important to Valve and Icefrog as the pub scene. Some of the heroes scraping the bottom of the barrel in public games (like Wisp, Enchantress, Oracle, and Lone Druid) are popular and effective heroes in pro tournaments, while top heroes in pubs (like Omni, Necro, and Riki) are rarely used in pro games. In the case of the first, these heroes are often difficult to play or are only effective as part of a team. In the case of the latter, heroes that are powerful in pubs often have a weak spot that pros are better able to neutralize.

In past patches Icefrog nerfed the heroes who were dominating pro and pub games. Often he critically disabled heroes, dropping heroes from win rates of fifty-five percent to forty percent. A variety of other heroes received small buffs. With incremental improvements, a hero may go unpicked, until one day players realize how powerful they are. Occasionally, a hero would be reworked. They might have one or all four abilities totally redesigned. In some intangible way, the hero would be the same, but in practice players would have to create a whole new method by which to make the hero successful.

Other aspects of the game are changed as well; adjustments to the map, new items, and rewards for killing enemy heroes. But it all misses the point. What is the goal of the biannual changes to the game?

The most obvious answer would be, to develop a game where all heroes have an equal chance of winning, and all items are considered valuable. That is not to say that all strategies succeed, or every item works for all heroes. It means that no hero would be successful sixty percent or forty percent of the time.

This doesn't seem to be what Valve is aiming for. Every time a hero becomes overpowered they suffer like Icarus, by falling in flames. A more measured approach would be a steady stream of minor nerfs until they stabilize at a fifty percent win rate, but Valve doesn't do this. Any that rise too high are often sent to the low forties as a punishment. And heroes that already sit at a fifty plus win rate continue to receive minor buffs as long as they don't dominate the scene.

This method is about creating a rotating cast. Since there are over one hundred heroes Icefrog's goal must be to highlight heroes for a period of time, but then allow others to shine. Some must fall so others can ascend.

This is both the most exciting and annoying part of the game. Having a favorite hero that you've devoted time to perfecting, crushed into oblivion by nerfs is beyond frustrating. This is mitigated by the fact that most players have at least five heroes they enjoy playing. Its also exciting because someday you're favorite may be on top. And if there is a hero that is driving you crazy, you know it won't last forever.

Considering all this, the feelings the pro have, the ups and downs of patches, is undoubtedly amplified. The best of them make a living off of Dota, and yet, in a constantly changing game, the skills of an elite player in one year may only be second or even tenth best the following one. Unlike baseball where the rules are solid, and players can expect their performance to be unaltered (aside from the inevitability of the effects of age), in Dota the reality of the game transforms, leaving only those who can learn and adapt to remain on top.

Comments