When can you consider yourself sub-x?

Same idea, you can get unlucky. I get sup 13 averages of 12 or 50 sometimes. Here is how to decide what your average is:
Ask yourself "What would I expect if I did a solve?"
Me? 12.6ish

he probably hasn't, but you get the point

the occasional bad ao12 is fine, but if you're getting sup 13 ao 50s, you're probably not sub 13. For example, I still get bad 9.3 ish ao 12s about once a week, but the worst ao 50 I will get is about 8.8.
 
the occasional bad ao12 is fine, but if you're getting sup 13 ao 50s, you're probably not sub 13. For example, I still get bad 9.3 ish ao 12s about once a week, but the worst ao 50 I will get is about 8.8.

One interpretation could also be that you are not sub-x until you have officially achieved it in competition (a major factor as to why many people keep cubing). So you would still not be sub-12, and I (although it pains me to say), would not be sub-4 minutes for 3x3 blind

-Doug
 
This seems really stupid to me, because by that logic, I'm sub 10, which is laughable to me (I average 10.5-ish)
Also, those bad solves are part of your average, because everyone gets good and bad solves. Personally, about 6 months ago, I fell into the bad habit of deleting my bad times because I was frustrated with them, but then my "averages" didn't reflect my global average because the times were just my good times. This seems almost exactly like this.

That's true, everyone gets good and bad times, but if I do an average of 5 not in an average of 50 I will basically never fail that much. When I do 50, some of my solves are super terrible. I get 25 good solves (it isn't like they are all great) and then around 15 terrible solves that I would almost never get if I weren't doing an ao50, and then I finish off the average with decent solves. I think what causes it is that after around 25 I am halfway finished, so I want to continue the streak of good solves, but then I mess one up, I get frustrated, mess 10 more up, and then finish up the average decently (sometimes).

by deleting your bad solves do you mean just deleting them and not resetting your session?
 
the occasional bad ao12 is fine, but if you're getting sup 13 ao 50s, you're probably not sub 13. For example, I still get bad 9.3 ish ao 12s about once a week, but the worst ao 50 I will get is about 8.8.

this is silly because I get more sub 13 than sup 13 ao50s, and the sup 13 ones are where I'm not warmed up and/or I completely flop. Besides, they are usually like 13.0x averages and then I get more 12.8x and 12.7x averages. Would this person be sub-13 (these 10 ao50s)
12.91
13.05
12.78
12.84
13.01
12.89
13.11
12.83
12.79
12.98

This isn't me, but it is an approximation of me.
 
Don't reset your timers. Ao1000 builds up over time. That is your true global average because it show good and bad times over a long period of time.
If I do 100 solves every other day on average, it will take me 20 days, or around 3 weeks. That can be enough time to see some improvement, so it is imo too long.

agree, I'm curious if CStimer can do Ao5000 or 10000
5000 would take me around 3 months. If I don't improve in 3 months I'll be pretty frustrated.
10000 would take me around 6 months. When you get into such long averages, it isn't gonna just represent how you solve right now, it will be pulled up by solves you got several months ago.
 
If I do 100 solves every other day on average, it will take me 20 days, or around 3 weeks. That can be enough time to see some improvement, so it is imo too long.


5000 would take me around 3 months. If I don't improve in 3 months I'll be pretty frustrated.
10000 would take me around 6 months. When you get into such long averages, it isn't gonna just represent how you solve right now, it will be pulled up by solves you got several months ago.

Another good thing of long averages like those is that you can plot them and see the trend lines, which mark your improvement over time. I'd say the trend line of an Ao10000 would give a more than good sub-X approximation but I'm too lazy to keep track of such long averages, I just keep an excel file with Ao100 (which I usually do in 1-2 days). If I see that I have less and less sup-20 Ao100 I would start considering myself sub-20 which has not quite happened to me yet. So I would be globally sub-21 :(

I need to warm up for like 5-10 solves before my fingers start working fine. But just because, after warming up, I usually can easily get an 18.xx Ao5 or even Ao12, I don't consider myself sub-19, even tough I have several sub-19 Ao50s. Anyway, more than sub-X or whatever I consider myself a decent speedcuber than enjoys solving. No need to be Feliks to enjoy cubing.
 
Wow folks I really didn't expect this to stir such a great debate :) thanks for all your comments. I'm gonna be open and say I categorise myself as sub-16 but I realise that I can still get sub-17 averages of 1000. I've never done an average of 5000 or 10000 for that matter but it's probably not a bad idea :)
 
I would do the average of 1000, but I know that when I finish, it will have solves from when I was a little slower so it'll make my average more than it actually was...
 
Capture.JPG
Due to this graph of my last 10,000 plots of my ao12's its safe to say i'm sub 18, this is about my entire speedsolving career, i like to see the improvement over time
 
I may be a little hard on myself but, I see if I have 5 consecutive ao100 first. This qualifies me for next round. Then I compare my ao1000 which usually takes weeks. I usually do 2 ao1000 but these days I do another ao1000 for verification purposes.
So 500 solves where I should have sub-x ao100 consistently and then 2 ao1000 for qualifying myself sub-x and then another ao1000 for verification purposes. But mostly, getting sub-x ao100 consistently is a good sign that you are already sub-x
 
I will probably do it more presicesly when I get faster, but atm I'm sub-50, and I haven't ever done an ao100, but I feel confident in that I can sit down and do a solve in under 50 at least 9/10 times. Last time I did a sesion of like 20 solves, my worst solve was 48 something. Although some would say measure in a bigger scale, because lucky scrambles happen, but when I have to mess up OLL or PLL to get over 50 I'm actually not going to bother do an ao100 or ao100 to check if I'm sub-50. Also when I'm this slow an ao100 would take me like over 2 hours, but I'll definetly do ao100s to test when I get under 35.
 
I would say you're sub-x if you can get multiple ao100's below a certain boundary. I have gotten a sub-18 ao100, but I haven't gone back and done another one yet. I may be able to say I'm sub-18 now, but I think I should do a few more averages.
 
this is silly because I get more sub 13 than sup 13 ao50s, and the sup 13 ones are where I'm not warmed up and/or I completely flop. Besides, they are usually like 13.0x averages and then I get more 12.8x and 12.7x averages. Would this person be sub-13 (these 10 ao50s)
12.91
13.05
12.78
12.84
13.01
12.89
13.11
12.83
12.79
12.98

This isn't me, but it is an approximation of me.

You can see from your description what you have done, which is take your best average of 50 and say you are sub x+a little bit more.
I think it's when you think you are sub-x. Nobody else but you can know how fast solves feel compared to how they are, or how often you pause etc.
You are the only real judge of what time you are below. However, generally speaking I think sub-x global average would be getting sub-x 4/5 times.
In terms of large averages, it depends. I have a session with approximately 20,000 solves in it, and the mean is low 15. Now, I would say I am globally sub-14, but my large session says otherwise.
I think an average of 1000 spread out over however long it takes you to solve the cube 1000 times is the best you're gonna get. Ao100 is too small a sample to determine a GLOBAL average imo.
 
Use ChaoTimer. Start a new session for every event each month. And the end of a month, you will know your global average. Start again.
Seriously though, If your PB ao100 is x*24/25, then you're sub-x.
 
Back
Top