There’s a +2 in the formula to the right.
-
Follow our Instagram!
-
Curious about what we're currently doing about bots? Check out the latest updates from staff in this thread.
-
Welcome to Smogon! Take a moment to read the Introduction to Smogon for a run-down on everything Smogon, and make sure you take some time to read the global rules.
Resource ADV Beginner's Lounge - Rules/FAQ, Question & Answer
- Thread starter vapicuno
- Start date
nyatalia I’d reccomend joining an ADV discord server if you’re looking for a regular supply of discourse on some of the more intricate matters of the tier.
Below is a link to the official ADV discord, but there are plenty of other servers available if it doesn’t quite fit for you
https://discord.gg/2Rv8V7GNV
Below is a link to the official ADV discord, but there are plenty of other servers available if it doesn’t quite fit for you
https://discord.gg/2Rv8V7GNV
What's the probability of Porygon2 not getting double critted by Dugtrio?
Is it 5.33%?
For simplicity sake, let's say that Dug is jolly, and there's no sand. So Dug must land 2 consecutive EQ crits in order to kill a P2 spamming Recover.
Enemy Dug kills something with EQ, then I revenge trap with P2. So Dug has 15 EQ PP left.
Should I calculate by taking that there are 14 chances left for double EQ crits?
Chance for a crit: 0.0625
Chance for a double crit: 0.0625*0.0625=0.00390625
Chance to NOT get double critted: 1-0.00390625=0.99609375
Chance that all 15 EQs do not double crit: 0.99609375^14=0.946679585
Chance that there's a double crit within those 15 EQs: 1-0.946679585=0.053320415
(so around 5.33%)
(Sometimes I also spam recover with other mons, such as my defensive Celebi vs opposing HP fire Celebi, to avoid losing to crits. But I wonder how safe this really is when there's a lot of HP fire PP to stall, and a double crit would kill me.)
Is it 5.33%?
For simplicity sake, let's say that Dug is jolly, and there's no sand. So Dug must land 2 consecutive EQ crits in order to kill a P2 spamming Recover.
Enemy Dug kills something with EQ, then I revenge trap with P2. So Dug has 15 EQ PP left.
Should I calculate by taking that there are 14 chances left for double EQ crits?
Chance for a crit: 0.0625
Chance for a double crit: 0.0625*0.0625=0.00390625
Chance to NOT get double critted: 1-0.00390625=0.99609375
Chance that all 15 EQs do not double crit: 0.99609375^14=0.946679585
Chance that there's a double crit within those 15 EQs: 1-0.946679585=0.053320415
(so around 5.33%)
(Sometimes I also spam recover with other mons, such as my defensive Celebi vs opposing HP fire Celebi, to avoid losing to crits. But I wonder how safe this really is when there's a lot of HP fire PP to stall, and a double crit would kill me.)
As usual for this type of problem, we should work on the smaller problem and try find a general formula or a way to use a computer to do it instead. I'll assume your calcs are correct in that only consecutive double crits will kill.
For remaining PP = 1, the probability of a double crit is 0 trivially.
For remaining PP = 2, the probability of a double crit is 1/256 ~ 0.4% as calculated above.
For remaining PP = 3, the probability is Pr(double crit in 1st 2) + Pr(1st hit no crit & next 2 are) = 1/256 + 15/16 * 1/256, ~0.0075683594 or 0.75%.
Remaining PP = 4, Pr(double crit in 1st 3) + Pr(double crit ONLY in last 2) + Pr(1st is crit, 2nd no crit, then double crits) = 0.75% as above + 15/16 * 15/16 *1/256 + 1/16 * 15/16 * 1/256
And so forth -- this is an easy thing for a computer to do but tedious by hand.
Your quick estimate for PP = 3 would be
Pr(next 3 don't double crit) = 0.99609375^2
Pr(double crit) = 1 - above = 0.0077972412
So yours is an overestimate (more pessimistic than reality). It will obviously get worse as PP increases, but how much worse?
Not too bad actually!
Using https://statisticsfundamentals.com/calculators/coin-toss-streak-calculator/
We see that the probability is 5.06% compared to your 5.33%.
Edit to add: I also verified my calculation for PP = 3 and PP = 32 (for HP Fire) has a double crit probability a touch over 10%.
For remaining PP = 1, the probability of a double crit is 0 trivially.
For remaining PP = 2, the probability of a double crit is 1/256 ~ 0.4% as calculated above.
For remaining PP = 3, the probability is Pr(double crit in 1st 2) + Pr(1st hit no crit & next 2 are) = 1/256 + 15/16 * 1/256, ~0.0075683594 or 0.75%.
Remaining PP = 4, Pr(double crit in 1st 3) + Pr(double crit ONLY in last 2) + Pr(1st is crit, 2nd no crit, then double crits) = 0.75% as above + 15/16 * 15/16 *1/256 + 1/16 * 15/16 * 1/256
And so forth -- this is an easy thing for a computer to do but tedious by hand.
Your quick estimate for PP = 3 would be
Pr(next 3 don't double crit) = 0.99609375^2
Pr(double crit) = 1 - above = 0.0077972412
So yours is an overestimate (more pessimistic than reality). It will obviously get worse as PP increases, but how much worse?
Not too bad actually!
Using https://statisticsfundamentals.com/calculators/coin-toss-streak-calculator/
We see that the probability is 5.06% compared to your 5.33%.
Edit to add: I also verified my calculation for PP = 3 and PP = 32 (for HP Fire) has a double crit probability a touch over 10%.
Last edited:
Thank you.
I think I'll be happy to continue with the plan of stalling out all of Dug's EQs. It's safer, and I end the fight at high hp (even though it costs 16 recover PP).
Hidden Power has 24 PP so double crit probability is 8.17%.
In desperate situations I've also used +1 CM Celebi to stall out Mixed Mence's fire blasts with recover. A crit would kill me, but given fire blast's 85% accuracy and 8 PP, this is a sequence that I've gone for at times. Now that I know about this calculator, I've learnt that the chance of a crit in 8 fire blasts is actually 35.38% (I input the crit probability as 5.3125% since fire blast accuracy is 85%).
I think I'll be happy to continue with the plan of stalling out all of Dug's EQs. It's safer, and I end the fight at high hp (even though it costs 16 recover PP).
Hidden Power has 24 PP so double crit probability is 8.17%.
In desperate situations I've also used +1 CM Celebi to stall out Mixed Mence's fire blasts with recover. A crit would kill me, but given fire blast's 85% accuracy and 8 PP, this is a sequence that I've gone for at times. Now that I know about this calculator, I've learnt that the chance of a crit in 8 fire blasts is actually 35.38% (I input the crit probability as 5.3125% since fire blast accuracy is 85%).
Last edited: