The early rounds of a map provide a complex challenge that is unique from the other rounds of a map. Since neither team has an established economy, the economic decisions made have perhaps an undue influence compared to later rounds. Additionally, while the early rounds may seem like “cheap” rounds, performing well in the early portion of a map has a significant effect on your likelihood to win the map.
The first round of a half is simple, since each side has $800 on each player, with only utility (or defuse kit if CT) decisions to be made. Additionally, the loss bonus is guaranteed for that first round: $1900. The outcome of this “simple” round actually creates a fairly large dichotomy in decision making for round two, however, for the team that loses round one. The team can either
I want to evaluate which method results in a better map winning percentage. It is also worth noting that there are some confounding factors, which I will address with how I will deal with it:
The data for this research comes from HLTV. All matches from the last 12 months were included. This time frame was chosen because it contained maps from the same map pool, as well as hopefully the same “meta”.
This resulted in data from 13408 maps. Removing the 2744 maps with a round 1 T loss with bomb plant left 10664 maps. It is important to note that this research may be invalidated by future meta shifts, map pool changes, etc.
The methodology for this is fairly simple: compare the map win rates for eco and force buys in round two for those teams that lost in round one. Given the fairly large sample size, traditional statistical methods may not provide very meaningful significance results.
The cutoff spend value to differentiate between a save and a force buy is defined at 4000.
With the designated cutoff value, there were 6617 force buys and 3981 saves.
Of the 6617 force buys, the map was won 2823 times and was lost 3794 times, for a win rate of 42.66%. Of the 3981 saves, the map was won 1418 times and was lost 2563 times, for a win rate of 35.62%.
Given the sample size (despite this really being more of a population-level analysis), we can infer that the force buy win rate is statistically significantly higher than the save win rate. Just to check, we can run the test:
##
## 2-sample test for equality of proportions with continuity correction
##
## data: c(2823, 1418) out of c(6617, 3981)
## X-squared = 51.082, df = 1, p-value = 4.429e-13
## alternative hypothesis: greater
## 95 percent confidence interval:
## 0.05423947 1.00000000
## sample estimates:
## prop 1 prop 2
## 0.4266284 0.3561919As expected and shown statistically, the win rate for the force buy rounds is significantly higher than the win rate for the save rounds.
After losing round one, a team is presented with two options: immediately force buy and attempt to break the oppositions economy, or save and attempt to build their own economy. This small research has shown that pursuing the force buy strategy provides a significantly better chance to win the map. Pursuing the round two save strategy implies also saving again in round three so that a full buy can be completed in round four. This research shows that the more agressive approach is more effective. This finding implies that the benefits of breaking the opposition’s economy outweigh the risk of damaging your own economy.
Some limitations of this research are not taking team quality into account and judging the outcome of a map on the first two rounds only. It stands to reason, however, that early round strategy does ultimately play a role in map outcome.