I'm a data-driven kind of guy, so I keep careful records during PVP. I've been asked to share the data I collected so here it is. Some folks asked last tournament and I never had a chance to pull it together, but I had some free time today and figured this would be a nice place to keep a record.
For this last tournament (Spiral), I used two teams on offense and several on defense in a desperate attempt to find one that worked. On offense, I started with a Havok/Captain Britain combination that worked wonders up until I reached Vib. I then ended up with only a 70% winning percentage in a trailing-10 average, so I switched to Havok/Iron Patriot (blaster).
For gear, my Blaster Agent carried: Hoarfrost, Neurotrope, Scroll of Angolab, and Quantum Jumper.
The best defensive team I managed to find was Swashbuckler Nightcrawler (inf) with Juggernaut
Defensive gear: Bruiser Agent, Signpost, Hotshot (reforged at 300), Quantum Jumper, Neurotrope
I finished the tournament comfortably in Vibranium (0.8%) but stuggled to gain ground during the last couple of hours as I tried to push into ADA.
What follows is a summary of the data I collected during the tournament and some of the conclusions I draw from them. If there is anything else folks are curious about, leave me a comment and I'll try to add it.
For each attack I made, I captured the following:
- Name of player
- Rating difference (negative when lower than my current rating)
- Level difference (negative when lower than my current level)
- Relative Armory Bonus (choice between Me, Them, or Equal)
- Relative Hero Bonus (choice between Me, Them, or Equal)
- Agent Class
- Gear Used (four slots for data capture; Unknown when I didn't see it or its effects)
- Heroes Used (two slots)
- Result (Win/Loss)
- Difficult (Yes/No; all Losses are automatically Difficult=Yes)
In the Notes section I kept track of things like horrific turn-orders, massive armory bonuses (200k+) coupled with minimal hero bonuses, players that just Outclass me in both bonus areas, and especially annoying combinations to fight against.
For each defense, I attempted to track the same data, but the only way to do that is to pull from what I learn on the attacking side, so I don't trust the last 6 data points to be accurate. Most players switch out between offensive and defensive teams, so it's likely they used different combinations when attacking me than I faced when attacking them. The data from the AFK fights are not included in the analysis below.
As mentioned above, this was my first tournament played while at level 300, and the results I got are much worse than any of my prior tournaments. The win percentage when on Offense is about 10% lower than it had been before, and the Defense % about 6% lower. That being said, I managed to finish comfortably in Vib (0.8%).
Within the 720 offense battles, I faced 260 unique players. The most attacks against any one player was 10 (did that twice) but the majority were one-offs early on in the tournament as I was climbing up to <1%. Once there, I faced the same 50 players repeatedly.
|Attribute||# Same/Higher||% S/H||Avg||#L||%L||Avg|
|Armory Bonus||317 Same / 319 Higher||44% / 45%||N/A||84||22%||N/A|
|Hero Bonus||76 Same / 387 Higher||11% / 54%||N/A||257||36%||N/A|
The above summary helps explain how I was comfortably in VIB while winning only 83% of my offensive battles. Prior to level 300, I was matched against players with a lower rating and level far more often than happened this tournament. That meant that while I didn't win as often, I won a greater number of rating points on average from each fight.
While it's not in the table, the same effect in reverse helped me in AFK - I faced fewer AFK players with lower ratings than usual, so I lost less each fight on average.
My other conclusions from the summary are that both of my bonuses are a bit a bit below average for placing in VIB/ADA. To that end, I've decided to dedicate the 10 gold I win in a tourney to boosting my armory at the start of the next. Adding and leveling heroes is a given, so no behavioral change there.
|WC Phoenix (blast)||135||19%|
|BS Quicksilver (gen)||100||14%|
|BS Quicksilver (scrap)||84||12%|
|WWII Cap (tact)||75||10%|
|Iron Patriot (blast)||74||10%|
Here are the heroes I encountered in at least 10% of my attacks. I have both "any version" and discrete versions tracked for heroes like Phoenix, Quicksilver, War Machine, and Cap as they tend to show up in a variety of forms.
Phoenix still dominates the meta (been this way for the last 5 tournaments), but Quicksilver made a rapid rise with the Blue Suit. This data (of course) influenced my choice of offensive teams and gear.
Big change from the previous tournament here. Scrapper went from the second-most common class to the least seen. I guess everyone was scared of Enraging Juggs. Infiltrator had a significant drop-off as well going from 20% to 12%. I blame BSQS for that.
|Scroll of Angolab||301||42%|
No real surprises here, except perhaps Neurotrope jumping Warbringer. In many cases, I finished battles too quickly to determine the complete gear carried by an opponent, so there are a ton of "Unknown" entries in my spreadsheet. I consider that a good thing :)
|Bad Turn Order||19||15%|
This is actually encouraging. Bad Turn Order, Team-wide Stun, and Crashes just happen and there's nothing you can do. The players that Outclass me or have Massive Armories are beatable, but only if luck is on my side. Really, only the ones that are Unclear are addressable through tactics or strategy and most of those occurred before I changed from my Havok/Captain Britain team to Havok/Iron Patriot.
This explains the 83% win percentage - I faced far more opponents that simply outclassed me than I did before reaching level 300.