Beyblade testing: didn't we miss something?

Attack VS Attack testing should be a staple part of testing a new attack combination.

I know it can be tedious and frustrating to conduct attack vs attack testing with some of the wins that occur from the testing simply being based on luck. But surely if an attack combination were to achieve a significantly higher win rate over another attack combination those results achieved must be an indication of the winning combinations strength?

Though in regards to Pegasis I remember Momo stating a theory he had one time that seems to correlate with my current testing of Pre- HWS. Momo stated, "In attack vs attack battles Pre-HWS Beyblades seem to come off second best against HWS due to the plastic of the HWS being able to absorb a part of shock that occurs during impact in battle". In my testing of Pre-HWS Beyblades so far Pegasis and Quetzalcoatl have both lost quite easily to their HWS counterparts such Vulcan, Lightning, Gravity, Meteo and Ray.
(Dec. 11, 2010  1:19 AM)Fyuuor Wrote: Attack VS Attack testing should be a staple part of testing a new attack combination.

In my testing of Pre-HWS Beyblades so far Pegasis and Quetzalcoatl have both lost quite easily to their HWS counterparts such Vulcan, Lightning, Gravity, Meteo and Ray.

Hmm... I think this should not be included in attack type testing. Attack types will win based on luck, and the launches are extremely sensitive to timing, and the position of launch and previously launched bey in the stadium. In order to keep results consistent, not only does one have to replicate these conditions (time, placement, power) but also make sure that these conditions are not giving one bey the advantage. I simply think that it is oo much work. Wouldn't it be more accurate to test attack types against defense and when the occasion DOES rise for attack vs attack, it will be experience and skill that will come into play. There are simply too many variables to control for testing.
At most I think Attack against Attack should be secondary testing, with everything concerning Attack against Defense done first.
(Dec. 11, 2010  1:26 AM)Kai-V Wrote: At most I think Attack against Attack should be secondary testing, with everything concerning Attack against Defense done first.

The question is can this even be tested to give consistent results for the majority like attack vs defense?
(Dec. 11, 2010  1:27 AM)Nojo294 Wrote: The question is can this even be tested to give consistent results for the majority like attack vs defense?
Hm, if you can at least shoot one before the other for ten rounds, then shoot the other one first for ten rounds, or even find another player and exchange Beyblades after ten rounds, it could still prove something in my opinion.
I kinda have to agree with nojo294. While there are many situations where there is attack vs attack, there is so many Little things that can affect these battles. The rate at which attack types with rubber tips RPM's slow, makes the amount of time between launches a huge factor. There is times that, when you launch an attack type, then when you launch the other one, attack type#1 is right at the one you just launched, and it is an instant KO.
What I'm trying to say is that small things like that could change the outcome in a way that
wouldn't be a factor in a competitive situation.

Although, these tests could be done with 2 bladers of approx. Equal skill and power(ex-not a 10 yr. Old with a 20 yr. Old). Another thing that could be considered is alternating which beyblade is launched first.

I do think that if there are to be attack vs attack tests, more rounds need to be done than normal tests such as- stamina vs stamina, attack vs Defence, Defence vs attack.

Edit:Kai-v stole a piece of my answer, while partially beating me at the same time.
Kai - V is right, Attack testing should be secondary but that doesn't mean it shouldn't be included at all.

Even with all the variables involved in Attack Testing, if an attack combination achieves a 80 -90% win rate over another Attack combination and is consistent in other members testing are you telling me that these results are meaningless and should be disregarded?

To exclude amount of luck and other variables Attack Testing should be over a larger amount of rounds to exclude variables. Possibly 20 rounds?
I also agree with 2 bladers battling with equal skill and power there won't be very different results and the testing would be more accurate.
30 rounds sounds good.
@ fyuuor-
IMO it is really an optional test that can be done; like kai-v said, it is a secondary test. I think that if there is a high win rate against a Defence type, and a bad win rate with that attack combo against an attack type, the combo shouldn't be considered less effective than it was before the attack vs attack results. Example- if there are two attack types that both achieve the same win rate against the same Defence type, but when they battle together, one has a higher win rate than the other. I wouldn't consider one a better choice for an attack type. That is just my point of view though.


And I was also thinking a little more than just 20 matches. Maybe 25-30, depending on the condition of the rubber tips.
(Dec. 11, 2010  1:45 AM)Fyuuor Wrote: Even with all the variables involved in Attack Testing, if an attack combination achieves a 80 -90% win rate over another Attack combination and is consistent in other members testing are you telling me that these results are meaningless and should be disregarded?

If this is the case yes, then that IS something significant. I'm trying to say however, will every member be able to control all the variables, and even if they do, will it be controlled in the same manner? For example, launch position, angle, speed, power all make a difference. This is especially important as none of the members will have the same preferences, physique and even stadium placement for that matter. However fyuoor, IF despite all these variables the majority of members achieve similar win rate, then yes it should be considered. But is it worth trying at all? I mean there are discrepancies on Defense types between members... how much worse will attack types be?
(Dec. 11, 2010  1:45 AM)Fyuuor Wrote: To exclude amount of luck and other variables Attack Testing should be over a larger amount of rounds to exclude variables. Possibly 20 rounds?

IMO all testing should be at least 20 rounds. I'm not sure why people usually only do 10 rounds; if only 10 rounds of testing is done with something, you are always going to need more tests to verify those results, so you might as well do 20 from the beginning.

Anyways, yes, as Kai-V said, Attack testing should be something secondary. And as long as the Bladers switch Beyblades half way through, that should be enough to minimize the effect of the variables, and to ultimately prove something. I don't think I have personally ever discounted it, it is just difficult to do when you don't have someone on a similar skill level with you all the time.
We need to start doing tests with the duotron launcher, seriously.
(Dec. 11, 2010  1:56 AM)SSJfisherman Wrote: Example- if there are two attack types that both achieve the same win rate against the same Defence type, but when they battle together, one has a higher win rate than the other. I wouldn't consider one a better choice for an attack type.

I'm sorry to say this but that view is incredibly naive. If both combinations both have the same win rate against one attribute but different against another the obvious choice is to choose the combination with the better win rate.

------------------

(Dec. 11, 2010  2:16 AM)Nojo294 Wrote: I'm trying to say however, will every member be able to control all the variables, and even if they do, ...will it be controlled in the same manner? For example, launch position, angle, speed, power all make a difference.

These sort of variables will always occur no matter whether it's Attack, Defence or Stamina testing.

(Dec. 11, 2010  2:16 AM)Nojo294 Wrote: But is it worth trying at all?

Yes of course. Just because the idea seems a little tedious doesn't mean it's not worth the effort. MF Pegasis 145RF is considered one of the best Attack type combinations currently available in the MFB meta game though in Attack VS Attack battles it performs horribly, perhaps achieving one of the worst win rates in Attack VS Attack testing. By performing these sort of tests it allows you as a competitive blader to make the appropriate decisions in battle for which combinations to use and gives you a rough guide to how the battle will pan out.

(Dec. 11, 2010  2:16 AM)Nojo294 Wrote: I mean there are discrepancies on Defense types between members... how much worse will attack types be?

No worse then Defence and Stamina testing if they're conducted properly like they should be.

----------------------

(Dec. 11, 2010  6:19 PM)Kei Wrote: IMO all testing should be at least 20 rounds. I'm not sure why people usually only do 10 rounds...

Perhaps it's laziness.

(Dec. 11, 2010  6:19 PM)Kei Wrote: Anyways, yes, as Kai-V said, Attack testing should be something secondary. And as long as the Bladers switch Beyblades half way through, that should be enough to minimize the effect of the variables, and to ultimately prove something. I don't think I have personally ever discounted it, it is just difficult to do when you don't have someone on a similar skill level with you all the time.

I agree it can be difficult to conduct proper testing but I think at least a few rounds against MF Lightning L Drago 100RF/R2F should gives us a true understanding of a combos potential, after all with the definition of a 'Balance Type Beyblade' changing Attack VS Attack testing might become a necessity to prove a combos true worth.
Given the weight of enough statistical evidence we can compute a margin of error... I realize that one can skew statistical evidence, but if there was enough testing, a pattern would easily be visible.

YES it is difficult to test, I think everyone agrees on that point, but Beyblade is a communal hobby. I cannot think of a more enjoyable way to pass the time then to blade with like minded individuals... (Whilst recording the stats of course) I think it is a MISTAKE to leave this area undocumented.

(Fyuuor I agree with MF Pegasis 145RF doing poorly, I have done a lot of testing against a Ray Gil 100RF...)
(Dec. 11, 2010  7:07 PM)Fyuuor Wrote: No worse then Defence and Stamina testing if they're conducted properly like they should be.

so ur saying that we should state the exact time intervals between launches, stadium position, and relative angle for the launches as well? This would be the only way for tests to give consistent results. If so, then I propose that attack testing be secondary, and there should be a standard for each of these factors.
EG. 2 sec between launches, wall on opposite side, 1/4 revolution at ~15 degrees full power.
(Dec. 11, 2010  10:04 PM)Nojo294 Wrote:
(Dec. 11, 2010  7:07 PM)Fyuuor Wrote: No worse then Defence and Stamina testing if they're conducted properly like they should be.

so ur saying that we should state the exact time intervals between launches, stadium position, and relative angle for the launches as well? This would be the only way for tests to give consistent results. If so, then I propose that attack testing be secondary, and there should be a standard for each of these factors.
EG. 2 sec between launches, wall on opposite side, 1/4 revolution at ~15 degrees full power.

If we wanted to be completely carp about it then yes. But getting members to work all that out while there testing is far to much trouble. We need an accepted and recognized set of rules which are easy to follow which are similar to how we conduct Stamina or Defence testing.

Perhaps due to the frustrating nature of Attack VS Attack testing it could be 20 rounds of a straight shot into the stadium with alternating launches as opposed to performing perfect sliding shot for both combinations while trying to achieve a perfect flower pattern while trying keep both combinations inside the stadium and avoiding self KO's will be extremely difficult.

If there is no definitive winner at the end of the 20th round with say a 65 - 70%+ win rate then testing should continue until 30 rounds where it will cease and we'll assets and evaluate the information.
(Dec. 12, 2010  2:16 AM)Fyuuor Wrote: If there is no definitive winner at the end of the 20th round with say a 65 - 70%+ win rate then testing should continue until 30 rounds where it will cease and we'll assets and evaluate the information.

I think that there should be standard conditions for attack vs attack testing, but if a trend does not appear in the first 10-15 rounds, its not worth mentioning. After all if one thing is superior to another a trend should somewhat be visible by the first 10 if not 15 rounds. If it is 50/50, then 20 rounds will suffice in my opinion.
The reason why I said 20 was due to a portion of the outcome of Attack VS Attack battles being based on luck. If the course of the testing lasted 20 rounds then that would reduce the amount of variable of luck as much as possible.

I still continue to support the idea of Attack VS Attack but only as a secondary source of testing. Perhaps the time to conduct Attack VS Attack testing would be if a new wheel or combo has an amazing win rate and is comparable to Top Tier status; performing Attack VS Attack testing could be an option to include as additional testing to help improve the case for the new wheel or combo.
Yes, I know, I'm pretty much bringing this back from the dead, but I was just thinking something that might be taken into consideration. (I was pretty hesitant to necro this though)

So, this discussion is talking about attack vs attack.
When it is attack vs attack, there are KO's, and some wonky ones too. but quite a bit, as described by yamislayer in the OP, is that a combo such as MF Pegasis 145 rf can KO a defense type, but can't outspin other attack types. Yamislayer is a super knowledgeable guy, and knows tons about attack vs attack, so if he says that you can easily outspin by using a slightly weaker launch, then that would reduce the amount of wonky Ko's

But, I really haven't been seeing any attack vs attack tests! Is it because people are not confident enough with their launching skills, or just plain old forget about it? How can we make this easier?

So, it has been discussed in here about attack vs attack being secondary testing. If we do what yamislayer says, which reduces the amount of KO's, then it comes down to an outspinning situation. When attack goes against attack, there will be a lot of times where the two beyblades make no contact at all for a good while, and only hit when they are almost completely out of spin.

This is starting to sound quite a bit like solo spin to me. Now this comes back to "Beyblade Testing: didn't we miss something?", And that there has not been people doing attack vs attack tests.

If people were to do solo spin tests with their combo(obviously attack vs attack would be preferred, and be the first option), then we can have a good idea of how well that combo will do in attack vs attack. It is also good so that people who are unsure of their launching skills to have more accurate tests.

If the solo spin times match the win % an attack combo gets against another attack combo, then it is something that I think all attack combos should include.

In case my point wasn't exactly explained well, here's an example(just making up numbers):

MF Lightning l Drago CH120 RF vs MF Vulcan Horuseus 85 RF
Lightning L drago: 17 wins
Vulcan Horuseus: 3 wins

Solo spin MF Lightning L Drago CH120 RF
10 rounds, 45 seconds average

Solo Spin MF Vulcan Horuseus 85 RF
10 rounds, 31 seconds average

------------------------------------
So lightning l drago does quite a bit better in both set of tests, and the results match. That would mean that it is ok to post solo spins for attack types. Once again, attack vs attack would be preferred, but for some members that are not super great at launching, solo spin may be more accurate.

But obviously, the Rf or whatever attack bottom is to be used needs to be in the same condition. As was described to me before by yamislayer, a worn Rf has more stamina than a new-ish rf. Heck, even use the same Rf
hmmmmm perhaps but that means for attack versus attack people always go to buy2 new storm pegasis, solo spin time would be better becoz then u can switch the RF
But wouldn't burn cancer have a better solo spin but at the same time gets defeated by MF lldrago ch120 rf.
ATTACK :
Well okay, we know Attack has TERRIBLE stamina and what if the the attack type bey didn't knock the other bey out the ring ? So why NOT use Metal Flat. It may not go fast but it can outspin maybe defensive type bey and it could knock out a stamina type. So test this combo MF Vulcan Horseus 85/90/100 MF against Top Tier ATK,DEF, and END types.

DEFENSE :
Well Defense has Stamina issues and 9 out of 10 Stamina beats Defense. So use PD or SD with with the following 2 wheels : Libra or Basalt. It's a HUGE RISK because these two bottoms that I told you about doesn't grip the stadium well at all especially PD !!! So test these combos MF-H Libra/Basalt Bull GB145 PD/SD against all top tier types.

ENDURANCE/STAMINA :
Well Stamina types just too Light !!!! Attack types are there weaknesses. So try putting A heavier wheel on it. Gravity, Libra, or Basalt. So try this combo Gravity/Libra BD145/230/100 WD/SD/PD !!!
Also if you want to Put a Metal Face on it because the Performance Tip doesn't grip well on the stadium !!!

JUST TRY IT AND TEST IT AND POST A LINK OR VIDEO
basalt may work with WD or PD in stamina and defence and welll MF has been proven work best with left spinners
Ok well Basalt is somewhat unbalanced so the only way to balance it is GB145 so I chose Libra/Gravity as an alternative. So MF will would do best with Gravity/MLD/LLD so that's good