Yes, if you look under 'Other' in Captcha Breaker and scroll down, you will see it solves 12%. If you set SER to give captchas 2-3 retries, that % will obviously go up, but not by that much more.
If it solves captchas 12% of the time, on average this is what you could expect from multiple attempts:
If the odds of success are 12%, that means the odds of failure are 88%, and for each attempt after, the same, since they are independent of each other.
Mathematically, the odds of failure would be represented like this for 2 retries (meaning 3 total attempts): 0.88 * 0.88 * 0.88 or you could also say 0.88 to the power of 3
Using that example (3 total tries), you'd get 0.681472, which is 68.1472% odds of failure, which means 31.8528% chance of successfully breaking the recaptcha.
Mathematically, here are the odds of breaking the recaptcha based on the number of attempts:
12%
22.56%
31.85%
40.03%
47.23%
53.56%
59.13%
64.04%
68.35%
72.15%
However, that doesn't mean you should crank up the retries to beat those recaptchas, there are also other factors at play here. For starters - not all recaptchas are equal, and the more you hammer them with your proxies, the more likely you are to start getting the dreaded double blobs. That 12% success is an AVERAGE across ALL recaptcha types. If you're hitting double blobs constantly, that number is going to be much closer to 0. Plus, more retries slows down the whole process.
I have to admit that @almostbasic hat is all right here. The problem with recaptcha is that it is giving you a more harder captcha if you failed to deliver a good result or even if you download a captcha too often from the same site without answering it at all.
+1 @almostbasic - Statistics was my favorite class - you gotta love the above explanations as they are spot on. And more attempts seriously degrades the next probability of success which is what I was getting at.
@mattb348 - CB can only do so much, but it is better than nothing (especially if you are on a budget). If you plan on addressing recaptchas in a serious manner, you need other services.
I tried Captchatronix and had pretty awesome results - you just have to make sure you set the timeouts in Captchatronix right, or you'll start getting double blobs all the time. I ran everything into CB, which was set up to only send recaptchas to Captchatronix, and it worked quite well. That was about a week ago that I finished my 1 month test, and I was building links like crazy. I'm still testing out a few other things, but so far I think I'll end up back there. It works out to be a LOT cheaper than DBC when I ramp the links up like that.
@mattb348 - Captchatronix is an OCR service (not humans) and (for the moment) ONLY supports recaptcha - so make sure you set it up to ONLY send recaptchas to it. (Very easy with the all amazing GSA CB). However, that's really all I needed - CB covers so many captcha types for me, there's not many types of sites I can't post to, except the recaptchas.
Comments
If the odds of success are 12%, that means the odds of failure are 88%, and for each attempt after, the same, since they are independent of each other.
Mathematically, the odds of failure would be represented like this for 2 retries (meaning 3 total attempts):
0.88 * 0.88 * 0.88
or you could also say 0.88 to the power of 3
Using that example (3 total tries), you'd get 0.681472, which is 68.1472% odds of failure, which means 31.8528% chance of successfully breaking the recaptcha.
Mathematically, here are the odds of breaking the recaptcha based on the number of attempts:
However, that doesn't mean you should crank up the retries to beat those recaptchas, there are also other factors at play here. For starters - not all recaptchas are equal, and the more you hammer them with your proxies, the more likely you are to start getting the dreaded double blobs. That 12% success is an AVERAGE across ALL recaptcha types. If you're hitting double blobs constantly, that number is going to be much closer to 0. Plus, more retries slows down the whole process.
I have to admit that @almostbasic hat is all right here. The problem with recaptcha is that it is giving you a more harder captcha if you failed to deliver a good result or even if you download a captcha too often from the same site without answering it at all.