Why would Growl and Tail Whip fail in Gen 2?

Question

I'm playing Gold on Virtual Console for my first time, and I notice that for some reason Growl and Tail Whip fail a lot when the opposing Pokemon uses it. Not miss, just fail. One time it happened when I was switching in, but the others were just normal turns.

Edit: A similar thing also happened where a Poliwag used Hypnosis on my Hoothoot immediately after it switched in. The move said it didn't affect Hoothoot, then the next turn Poliwag used Hypnosis again on the same Pokemon and it hit. I'm starting to think it might just be using different text boxes to represent the move missing. Hypnosis has 60% accuracy and according to many semi-reliable sources, there is a 1/4 chance for a status move used by an opponent in Gen 1 and 2 to miss.

Furthermore, this question is not a dupe, as the question that this may be a duplicate of has a questionable answer. Read comments below if you are confused.

Edit 2

Just in a battle Onix vs. Lickitung. Onix is my Pokemon. Onix attempted to use Screech twice. It failed both times. Third turn it works. There was nothing that would have prevented me from using Screech other than accuracy.

Becoming more convinced of the accuracy thing.

Comments

Read my comments above.

---

Yes they were not as low, one of the times it was immediately after I switched in

---

"Furthermore, this question is not a dupe, as the question that this may be a duplicate of has a questionable answer."
https://pokemondb.net/pokebase/meta/75536/repeat-questions-questions-that-unanswered-against-rules
You asked your question yourself. I believe this question still counts as a dupe, as the other are unresolved.

---

"Onix attempted to use Screech twice. It failed both times. Third turn it works."
Screech's accuracy is only 85. If it's saying it failed when, in reality, we've figured out why it's saying that failing and missing are the same thing in this instance, however, we still don't know *why* Growl and Tail Whip have missed.

---

I am aware. The text said the move failed and didnt say it missed. That's why I posted that. I know Screech doesn't have 100% accuracy.

I also asked that meta question after I made this one and before I was aware there was a similar yet incorrectly answered question

Answer 1

In Gen 2, moves selected by an AI have a 25% (64/256) chance to miss.

Source: Several websites like Twitter, gameFAQs which are normally unreliable sources but the answer seems to be the same each time

this, and this have the same answer as well

Comments

I personally wouldn't use a Twitter thread as a reliable source. I've looked quite a bit on Bulbapedia and I haven't seen anything about this, and I feel like this is something that would be included on the page for either status moves or Generation I, but I haven't seen anything about it. I'm not saying this information is wrong, but I'd highly recommend cross-referencing things that come from things like Twitter threads and Reddit posts and such.

---

If it says it failed, that basically missed and the status moves might not have as good as accuracy as they are in later games. The Pokemon's accuracy might also be lowered due to smokescreen or other moves.

---

I wouldn't either, however I didn't find anything else. I think I found something on GameFAQs but that's pretty similar to Reddit.

---

Growl and Tail Whip have 100% accuracy, and that has never changed. Failing and missing are 2 different things, and there is a different message for both so "If it says it failed, that basically missed and the status moves might not have as good as accuracy as they are in later games." don't know what you mean about that. The asker would've talked mentioned Smokescreen, but he didn't.

---

If you can't find anything else, that's a pretty good indication that it's not necessarily true. If this was something that happened I feel like it'd be pretty well documented.

---

This is the only think I could find that's somewhat similar to the situation you're describing, though it's for Generation I and not Gen II like the asker describes in the question.
https://bulbapedia.bulbagarden.net/wiki/List_of_glitches_(Generation_I)#1.2F256_miss_glitch
Furthermore, the links below have details on how the AI works in Generation I, and it does not seem to have any mention of the information in your source, as far as I can tell, although it may be incomplete.
http://wiki.pokemonspeedruns.com/index.php/Pok%C3%A9mon_Red/Blue/Yellow_Trainer_AI

---

I'd personally appreciate if you found and linked to all the places you got your sources. Not only should you anyway, but I'm personally curious about this and want to look more into it.
What I find really strange about this is that, at least, in my opinion, it'd be really dumb to intentionally code your game so that every trainer you fight will miss 1/4 of the attacks they use, which is a pretty massive amount and a huge handicap. Even if you're trying to make the game easier, I feel like 1/4 is far too much. Second, the number 64/256 keeps popping up. Why isn't is simplified to 1/4? I believe that 256 is a significant number for probabilities in older games, at least, from what I've seen, however, it doesn't really make sense why it's not just simplified to 1/4, as it's much easier to comprehend 1/4.

---

Why was I downvoted? A source like Twitter or something else is not that good but people there and on similar websites are having the same problem and the answer is almost always 64/256. One of my sources is not reliable but I’m confident this answer is correct.

---

"One of my sources is not reliable"

Most likely because of that.

---

I downvoted this because I'm personally not convinced that's what the answer is. I've spent quite a bit of time searching for it, and I've found nothing over AI acting this way. On top of that, the 1/4 vs 64/256 thing I mentions in my comment above makes me think that all of those answers came from the same source.

---

I couldn’t find the GameFAQs answer but I added some more answers that say pretty much the same thing

---

I still don't trust those, honestly, as they're unsourced and they still use 64/256, so I assume that they're from the same source, or something that leads to the same source.

---

The first source was posted 2 years before the tweet

---

It's still unsourced, and I've found nothing about this mechanic on non-forum websites. It's also about Gen I, rather than Gen II.

---

I think Gen 1 probably had similar battle AI to gen 2

---

Also, I was fighting a Poliwag and it used Hypnosis on Hoot hoot and it said it didn't affect it, then next turn it used it again and it hit? Why does this keep happening? Is it just using different text to represent the move missing?

---

As far as I've found, the only explanation for this is that the opposing Pokemon's accuracy is lower than 100 or your Pokemon's evasiveness is higher than 100

---

But would it say it missed or that it failed? That is the big question here

---

Yes, it should, but I'm personally finding it harder and harder to believe this "the AI has a 25% chance to miss an attack" thing as I've not come across many sources that mentions it.
I'll keep a list of sources that don't mention it below, so you can understand how much I've looked at to try and find support for this.
No mention:
https://bulbapedia.bulbagarden.net/wiki/Status_move (I'd expect Bulbapedia to mention this mechanic in at least one of these pages listed here, but so far no luck)
https://bulbapedia.bulbagarden.net/wiki/Accuracy
https://bulbapedia.bulbagarden.net/wiki/Generation_II
https://essentialsdocs.fandom.com/wiki/Battle_AI
https://www.math.miami.edu/~jam/azure/forum/buzz/ultimatebb.cgi?ubb=get_topic;f=1;t=001216;p= (Not sure what the original topic of this thread is, but it mentions moves that have 100 accuracy missing)

Support:
https://pokemondb.net/pokebase/295915/why-did-poison-powder-fail?show=296058#a296058
https://pokemondb.net/pokebase/49221/why-do-status-moves-used-by-the-computer-enemy-fail-in-gen-i?show=259174#a259174 (both unsourced)
https://twitter.com/chuggaaconroy/status/1024355701962014720 (doesn't mention 64/256, and I've got no clue where this information came from)

---

Ok, but yes it should what

---

It should say that it missed, rather than saying it failed.

---

Ok, then I will BA. I know Bulbapedia doesn't mention it but I feel like if all of these other sources have the same answer then it's probably true. There is power in numbers.

---

https://bulbapedia.bulbagarden.net/wiki/List_of_glitches_(Generation_II)#Lock-On.2FMind_Reader_oversight
This is something I've found on status moves saying that they've failed instead of missing, though the situation is different from what you've described

---

There's much more power in the numbers of things that should mention it but don't. I've been looking into this for quite some time and I'm personally convinced that it's not true at all. It's like your friend in second grade who told you that pressing and holding A when you catch a Pokemon would make you more likely to catch it. That's just something that kind of spread around without any evidence, and I feel like this is one of those situations where somebody saw something and it just kind of spread. Obviously to a lesser extent, but lack of any evidence and what seems like lack of any mention on Bulbapedia, which, like I said previously, would definitely have this information if it was true, makes me believe that this is false.

---

Ok. I understand. So what should I do regarding the answers and the fact that this question is a dupe of a question that *probably* has the wrong answer

---

I don't think you can do anything unless you want to look for yourself, but you'll probably come up empty handed, as I've found nothing about it, not even the gamefaqs thread, or the "several websites" that this answer mentions, and I've been looking for at least an hour.

---

Ok, don't overwork yourself trying to find an answer to this question

---

https://gamefaqs.gamespot.com/boards/375087-pokemon-crystal-version/77202051
I think this is the gamefaqs thread the answer refers to. No numbers or source is mentioned, and the person who said it said "that's what I'm assuming," after somebody asked a further question about it, which tells me that they don't really have any proof of this. A post on this thread also claims that "it failed" it just alternative text for a move missing, so that may be why it says that it failed when it didn't really, however, without accuracy or evasion changes, it's strange that these moves are missing.

---

Another possibility is that your Pokemon's stats were already as low as they could go or that Mist was used, though I find that unlikely.

---

I'll give you a play by play of one of the turns this happened.

A wild Rattata appeared!
You sent out Buggy (Metapod.)
Turn 1
You switched to Hoothoot
The wild Rattata used Growl
But it failed!
Turn 2
Hoothoot used Tackle (takes damage)
The wild Rattata used Growl
But it failed!
Turn 3
Hoothoot used Tackle (takes damage)
The wild Rattata used Growl (lowers attack.)
???????

---

For anyone reading this in the future... this answer is (generally) correct. This mechanic was confirmed by looking through Gen II's source code, and it's now mentioned on Bulbapedia:

"In the Generation I and II core series games, when an in-game opponent outside of the Battle Tower uses a status move that would decrease one of the target's stats, that move has a 25% chance to fail in addition to its normal chance to miss. In the Generation II core series games, this condition also applies to status moves that would inflict a non-volatile status condition upon the target."

https://bulbapedia.bulbagarden.net/wiki/Status_move
https://github.com/pret/pokered/blob/master/engine/battle/effects.asm

---

Haha thanks Amethyst. (This thread has an absurd amount of comments )