For those not in the know, the Super Training feature exposes the "effort values" of a Pokémon to the player, allowing the player to increase each stat individually, or wipe them all clean using something called a Reset Bag. What's not obvious is that high-level Pokémon can use a Reset Bag to show its base stats sans-effort-points, which can then be used in reverse to determine the "individual values" for each stat. These are kinda wonky terms, but it basically means this:
• Every Pokémon species has a pre-determined set of "base stats" for each of the six combat stats: HP, Attack, Defense, Special Attack, Special Defense and Speed.
• Individual Pokémon have "individual values" (IVs) determined at time of creation, which are wholly random numbers between 0 and 31 that never change. A higher IV in a stat will yield slightly higher values in that stat than lower numbers.
• Each opposing Pokémon that you defeat in battle has what people call an "EV yield", which is from 1 to 3 "effort points" (EVs) for pre-determined stats. Multiple effort points can be awarded to the same stat by a single defeated Pokémon.
• The higher the EV for a particular stat, the higher the value for that stat. So as you battle with your Pokémon, its stats will increase even if it doesn't level up.
• A Pokémon can have 510 effort points total, in which case it is considered "fully-trained". Individual stats cannot exceed 252 effort points (previously 255, but anything over 252 didn't have an effect at level 100).
Super Training is all about the effort points. The Reset Bag shows you what the value was prior to deletion, which you can restore by restarting the game without saving... The implication here is that you can plug the stat numbers and EVs into a reverse formula to resolve the hidden IVs for your Pokémon.
I did some experimentation with this today and had much success. The Drifblim I mentioned earlier came off of the GTS, since I wanted the Exp. boost. Turns out it had a Bold nature, which is the absolute best for what I want to do: use it as an evasiveness tank (Minimize + Baton Pass). Bold makes the Defense stat 10% higher, but it also makes the Attack stat 10% lower. That's great, because Drifblim's base Defense is lower than its Base Special Defense, and its base Special Attack is higher than its Base Attack.
By doing some algorithmic voodoo, I ran a routine to crunch the numbers to make the Drifblim as robust as it could be in a general sense, using the following two parameters: 1) the final Defense and Special Defense stats should be the same to protect from any kind of attack, and 2) using the Defenses and HP stats, find the combination that will result in the least percentage of total HP lost by any given attack. That is to say, the combination where incoming attacks will hurt the least relative to the full HP bar. My program spat out the calculated EVs and the expected stat totals, and after Super Training was complete, it was spot on, save for HP, which turned out to be 1 higher in the game (meaning I probably got the wrong IV for that stat due to a rounding error).
I then did the same thing with a Shuckle. Gadzooks, Shuckle is a pain to train up using Super Training. But I got it to the point where it's as sturdy as it can get, and holy cow is that thing survivable. It was curious to see how the stat distributions wound up different from the Drifblim, due to a combination of base stat differences and a really pitiful roll of IVs on the Shuckle. Where the Drifblim didn't train its HP much at all (it needed 28 effort points in the stat), the Shuckle had to max it out with 252. And where the Drifblim trained both of its Defense stats most of the way up, the Shuckle has all of 6 effort points in Special Defense, with Defense at 252, and the final stats are the same. Just goes to show, there's variance between Pokémon!
In any case, it's a blast having access to all the numbers at long last, and I find myself "EV training" for the first time. Ever. |