Submitting my first post to the boards here, featuring a finished Doubles streak of 91 wins. Am I happy to contribute to this thread with a bare minimum wins streak? Sure, but I'm also going to try and improve on my mistakes and do this team justice! Without, further ado, here's the squad!
Pests In The Garden (based largely on Worldie's Flower-inspired Team)
![]()
First slot is Aromatisse @ Lum Berry
Ability: Aroma Veil
EVs: 252 HP / 180 Def / 76 SpD
Quiet Nature
IVs: Meh in Attack, 0 Speed
- Moonblast
- Psychic
- Misty Terrain
- Trick Room
You can look at Worldie's post here to get an idea of what this girl can do, but I'll just expand on a couple of points here. First, Lum Berry cannot be understated. Aromatisse will go last 9 times out of 10, so having the ability to shrug off sleep or paralysis hax in order to get Trick Room up turn 1 is huge. You'll also be using Trick Room 90% of the time on T1. The only time I ever really found myself doing something different T1 is if I know Aromatis will survive on T2, and there's a serious status threat that would for sure cripple the other active member. Thankfully Aroma Veil stops Taunt on both active members, which ensures outside of a flinch or OHKO that Trick Room will go up. Other than that it uses Moonblast or Psychic on the following turns depending on resistances.
![]()
Second slot is Incineroar @ Leftovers
Ability: Intimidate
EVs: 252 HP, 252 Atk, 4 SpDef
Brave Nature
IVs: 0 Speed
- Fake Out
- Flare Blitz
- Darkest Lariat
- Brick Break
The first difference from Worldie's team. Whereas they ran Hariyama as their Fake Out user, I wanted something with a little more bulk. Incineroar also has the added benefit of having Intimidate, which can be a huge boon against Mons that require a high roll to OHKO the front 2. Flare Blitz and Darkest Lariat are the obvious STABs, with Flare Blitz having the added bonus of a very high base power already, which gets plenty of unresisted KOs by itself. Leftovers is lopped on to help mitigate any recoil taken. Brick Break may seem like an odd choice here, and it definitely is. My reasoning for taking it over Low Kick or Superpower is that this team can easily get miss out on some essential KOs if the opponent sets up an early Reflect or Light Screen. Not enough that it's a consistent issue, but it's incredibly useful to have the option when the need arises. Also, having a consistent base power against lighter targets is handy as well. Overall I think this is the most flexible moveslot on the team.
![]()
Third slot belongs to Vileplume @ Grassium Z
Ability: Effect Spore
EVs: 252 HP, 252 SpAtt, 4 Def
Quiet Nature
IVs: 0 Speed, near max in SpDef
- Energy Ball
- Sludge Bomb
- Toxic
- Moonlight
This wonderful lady resists or can take a hit from everything that would threaten the frontline. It's also one of my favorite gen 1 pokes, so when I was looking for a team to build around this was the first thing that caught my eye. Dual STAB's are there to threaten a surprising amount of enemies, and Moonlight is there to help keep HP up in the few occasions when it's not being full aggro. Toxic is where I stray from Worldie's set. They run protect, but I wanted something to deal with especially bulky stall sets. Anything with excessive recovery that doesn't get KO'd in one turn by Plume and her ally can reliably stall out Trick Room, which leaves this team in a very bad place if Aromatisse is already gone. Even with max SpAtt investment and Z Energy Ball, I still found myself failing to KO or 2HKO some important bulky water types, which is an issue I might revisit later if I try for another long streak with this team. Honestly in my opinion she's the most fun member of the team to play around when I got the chance to swap her out.
![]()
Last but not least is Golisopod @ Assault Vest
Ability: Emergency Exit
EVs: 252 HP, 220 Att, 20 Def, 16 SpDef
Sassy Nature
IVs: 0 Speed, near max in SpDef
- First Impression
- Leech Life
- Liquidation
- Rock Slide
When everything else fails, throw Golisopod at it. My other large departure from Worldie's set. I don't think I EV trained correctly since looking back at him now some of the stats aren't where they should be, but this was the intended spread. Plenty of people have used and succeeded with him, so I'll just add my two cents. Emergency Exit is incredibly useful for creating space and controlling tempo, even moreso on a Trick Room team. First Impression is a good tool if he comes out after Trick Room expires, otherwise most of the time he'll be going for the super effective option. Getting forced to switch out when he has yet to move can be a pain, so if I run this team again I'll definitely look at damage calculators to make sure I know when to keep him in vs risking a swap.
Battle Videos
Battle 10 vs. Lass Chan
Shows how dangerous flying types can be against the backline. From what I encountered, Flying, Rock/Grounds with Sturdy, and bulky Water types were the most dangerous threats to this team.
Battle 58 vs. Veteran Dooley
Mawile after a Swords Dance is very hard for this team to deal with, luckily I dealt with the rest if the team fast enough. Also, not the first time I tried to use First Impression after Golisopod had been out for a turn already. Just sloppy play.
Battle 90 vs. Pokemon Trainer Cynthia
I was happy to see Cynthia for my 90th battle, she's one of my favorite trainers in the game. Not much to say about this one. Fake Out on Lucario triggers Steadfast which makes it even slower under Trick Room. Mega Garchomp comes in and could have been a big threat....had it not used Sandstorm. That triggers Emergency Exit which brings Vileplume in at the perfect time to finish off Togekiss. Cleanup from there.
Battle 92 vs. Pokemon Breeder Danby (The loss)
I got past 90 and started playing sloppier. A lot of RNG on this one on both sides. I should have switched Incineroar out sooner to the obvious Hydro Pump from Wishiwashi, but all I got out of him was Fake Out chip damage. Aromatisse and Drampa take turns slapping each other, which Aromatisse narrowly wins and manages to avoid a Hydro Pump. That's about when the game decided it had had enough. I must have rolled above average on Fake Out and Bloom Doom since there's a 87.5-103.9% chance to KO Wishiwashi through Rindo Berry on that. Rolled just high enough to drop it to single digits, and this is where it all went downhill. Rhyperior survives Energy Ball due to Focus Sash, and hits Plume with Payback, triggering Effect Spore, which of course causes paralysis, pushing Rhyperior to go first next turn. Aromatisse KOs Wishiwashi and enemy Golisopod comes out. Rhyperior manages to land Horn Drill onto Plume through the paralysis, and Golisopod lands Rock Slide onto Aromatisse. My Golisopod comes out as Trick Room ends. Enemy Pod lands a Rock Slide which flinches, sealing the nail in the coffin. I could have tried for First Impression onto Rhyperior, but I feel like it was still too far gone at this point.
Pests In the Garden QR code for those who want to try it (looking at you Smuckem)
Record of 147.
My streak finally got ended by me not clicking protect :( against a + Charizard X
Last Battle Battle Code: 9EQG-WWWW-WWX5-RH84
Is there any more Data I Need to provide to get on the leaderboard?
I’m looking at the pictures on my terrible tiny phone screen, so I can’t see the smaller details of your Pokémon. Do you remember the EV sets you used?
I see what I got wrong
Video of me losing made by me with my smartphone camera ;)
Sweet!
You might have to be patient because the mod has real life things to take care of, so I look forward to seeing you on the leaderboard when they check up on the thread!
Yes, but Kiele was pointing out that you won't immediately see your streak added to the board. This is because the moderators who will add you will get to it when they have a chance, but that won't be immediate since the moderators have other things going on in life.I just overread the part that I have to add a link to my video.
Isn't that a requirement to get added to the leaderboard?
While not "strictly required", main reason to provide the replay or video of the loss is to show that the streak is legitimate and not invented.I just overread the part that I have to add a link to my video.
Isn't that a requirement to get added to the leaderboard?
If you're going just for 50 i assume it's usable. Fast, high enough dmg and coverage to bruteforce prelegend sets.Is Mega-Aerodactyl worth using in the Maison?
I’m thinking of using it in Super Multis. The AI has Greninja-3 and Golisopod-3.
Also, how do you save battles before the loss? Save record option? Would have loved to have saved the one where Moltres and Kingdra kept setting opposing weather alternate turns!
I feel like necroing this to tell you that you can get bottle caps if you get a high enough streakYou get Battle Points.
Bottle Caps have nothing to do with the Battle Tree. They can be "bought" with shards or found while fishing. Some NPCs also give you Bottle Caps.
You have to lose the appropriately sized streak to get the rewards as well - you don't get them until that happens.I feel like necroing this to tell you that you can get bottle caps if you get a high enough streak
what are the odds of aroma4 stalling out these huge power iron heads with enough mist pp left to also deplete 16 play roughsQuick probability primer for Battle Tree
Pokémon is a game that occasionally rewards knowing (very) basic probability theory and having a calculator at the ready (your computer will do).
Before we jump to solving problems, we need a definition of a probability. Suggestion: a probability is a number between 0 and 1 (inclusive) assigned to an event as its chance to happen, where an event of probability 0 cannot happen, and an event of probability 1 is inevitable. Furthermore, to each event, there is its opposite ("success" -- "failure", or to the same effect, "happening" -- "not happening") and if an event has a probability p, the opposite event has the probability (1 - p). (This ensures that adding them gives 1 -- meaning that it's inevitable that an event either happens, or does not happen: "tertium non datur".)
As a fundamental rule, probabilities that depend on each other are multiplied, not added. If I flip a fair coin, the chance of heads is 1/2; if I flip it twice, the chance of only heads is 1/4, since the second toss depends on the first one landing heads.
(The "%" sign simply means "/100", thus 100% = 1, 80% = 4/5, and 23.81% = 0.2381; whether you express probabilities in percentages is arbitrary.)
Now we can jump to solving problems:* -- multiplication, / -- division, ^ -- exponentiation (raise to power), ! -- factorial, log -- logarithm (to arbitrary base; only in problem 3)
(1) Given n chances, what's the probability that an event, whose individual chance to happen is p, happens k times?
Concrete example: Across n = 5 turns on which it gets to act, what's the chance of an enemy holding Quick Claw triggering that item (exactly) k = 3 times? (The Quick Claw trigger chance is p = 1/5 = 20%)
Solution: We're looking for the result of
which is generally called the binomial distribution for n, p, k. In our concrete example, (5 choose 3) * (1/5)^3 * (4/5)^(5-3).Code:(n choose k) * p^k * (1-p)^(n-k)
(n choose k) is the binomial coefficient. To give a rudimentary explanation, it counts the number of ways you can choose k items from a pool of n total items. After all, we have to consider every possible ordering of three QC turns and two non-QC turns.
You can compute it as n! / (k! * (n-k)!) -- or, which is more convenient manually, from Pascal's triangle (named after Blaise Pascal, although he did not invent it first):
Start counting the rows at 0, then (n choose k) is the k-th number in the n-th row, also counting from 0. This means n is the 1st (not 2nd) number in the n-th row, whereas the 0th number is always 1, as is the n-th number. In this case, we need (5 choose 3) -- I've left out the 5th row as an exercise to the reader.Code:1 1 1 1 2 1 1 3 3 1 1 4 6 4 1 [...]
The chance of three QC activations in five turns is thus 5.12%.10 = 6 + 4; the complete row is 1 5 10 10 5 1, with the number that we want highlighted. Incidentally, note the triangle's symmetry.
If n and k are identical, e.g. you want to know the chance of n = 3 QC activations in a row (i.e. on k = 3 turns), (n choose n) = 1 (there is only one ordering) and (1-p)^(n-n) = 1, so you can simply calculate p^k; in the example, (1/5)^3 = (1/125) = 0.8%.
(2) Under "normal conditions", what's the chance of taking a freeze from an opposing Blizzard in Doubles? (Assume both targets will survive it.)
Solution: We have to consider multiple scenarios leading to the same outcome. Blizzard has to hit (7/10) and freeze (1/10) at least one target, which will happen if Blizzard
You could illustrate this as a (binary) tree of branching scenarios (targets named A and B):Code:(i) hits both targets, freezing at least (!) one (ii) hits one target, freezing it, and misses the other
The basic rule to determine the total probability with such a tree-model is that you multiply probabilities when going vertically through the levels of the tree (i.e. when events depend on previous events), and add them when you're looking at multiple branches on the same level. (In fact, the probabilities associated with the scenarios on any level will always add to 100%: e.g. the third level has 49%, 21%, 21%, 9% outcomes.) If you're puzzled, I'll show you in a minute.Code:100% ("Blizzard is used") 70% hits A 30% doesn't hit A 70% hits B 30% doesn't 70% hits B 30% doesn't 10% freezes A / 10% freezes B 10% freezes A 10% freezes B 0% freezes anything
Moreover, you can use inversions to pack several branches into one. For instance, on the path where Blizzard hits both targets, we could just look at the event where both hits fail to freeze, since we're interested in its opposite; if the chance to succeed is 1/10, the chance to fail is (1 - 1/10) = 90%. By the above rule regarding events that depend on each other, we have to multiply the chances of freeze failure. Thus, the chance of failing to freeze either target is (9/10)*(9/10) = 81%, which makes the chance of freezing at least one target (100% - 81%) = 19%, by the rule of opposite events' probabilities.
Ensure you understand that it is not 20%; don't simply add the probabilities of events that depend on each other. However, another correct way to obtain 19% is to consider the overlap between events: there's a 10% chance to freeze A, a 10% chance to freeze B, and a 10% * 10% = 1% chance to freeze both A and B. Indeed, 20% is wrong because you're counting the double-freeze event twice (it's included within the chance to freeze A, as well as within the chance to freeze B); subtracting its probability once corrects the error.
Generally, for sets of events X, Y,
Code:P(union of X, Y) = P(X) + P(Y) - P(intersection of X, Y)
Either way, for the aforementioned scenarios, we traverse the tree top-to-bottom, multiplying the probabilities:
We sum up the scenarios in accordance with the basic rule, because they are all leaves (or "end nodes") at the (same) bottom level of the tree. Thus we obtain 13.51%.Code:(i) 1 * 7/10 * 7/10 * 19% = 9.31% (ii) 1 * 7/10 * 3/10 * 10% = 2.1% -- count this twice, once for A, once for B Sum: 9.31% + 2 * 2.1% = 13.51%
Exercise: Assume the Blizzard user is Mesprit4 (@ Wide Lens). What's the probability of taking at least one freeze now? (The Wide Lens increases the accuracy of the holder's moves by 10% of its original value.)
14.8071% -- which is not quite a 10% increase from the previous result.
(3) Assume (arbitrarily) that every battle you fight with your Tree team has a 1% chance of loss; otherwise you win. What's the highest streak number that you have at least coinflip odds (50%) to obtain?
Solution: We're looking to solve (100%-1%)^n = 50%, which means log 0.5 / log 0.99 ~= 68.9, rounded down. That's 68.
Exercise: Same as above, but assume your chance to lose any given battle is 0.1%.
This does not mean you need exactly two tries to hit said number, but rather that after two tries, you have a 75% chance to have hit said number at least once (if the estimate for your victory chance is correct...).Then you have coinflip odds of a 692 streak.
End guide. If I've forgotten or screwed up anything, or it didn't help you (and you didn't know all this before), let me know.
(admittedly it was because we face Shirona (that Sinnou champion) + Kukui, running full Fairy set; His Primarina managed to OHKO my Mega-Mawile just when I thought I could laugh at her Spiritomb).