I read somewhere that there is something called split testing where I
keep changing two adverts when one of them has 30 or more clicks than
the other advert on the same group.

But I'm not sure how to to that.

In other words, lets say that one advert has 60 clicks and the other
has 30 clicks more (90 click). I delete the advert with the 60 clicks
and write a different one which hopefully will beat the advert with the
90 clicks. Now I don't want to delete the good performing advert, the
one with the 90 clicks, because it will stop my campaign and none of
the adds will generate impressions and clicks..

Question:
Is there a way to reset both adverts so both start with zero clicks
that way days later I can check which advert has the most clicks?

Make sure that you turn "optimized serving" off. You want both ads to
be served equally.

30 clicks is a good rule of thumb. A more accurate way is to use
statistical confidence. There's a handy statistical confidence
calculator for AdWords here;

    http://splittester.com/

You enter the # of clicks and CTR for each of the two ads, and it will
calculate a confidence level.

Changing just one letter of an ad will trigger a review. However, the
ad will continue to run on Google Search. It will stop running on
Search Network and Content Network.

It would seem to be a good idea to run your ad only on Google Search
until you have optimized it. If you run on Search Network, in
particular, you muddy things up considerably. First of all, you will
see a blended CTR that includes both Google Search and Search Network.
But Adwords uses only Google Search to determine quality score.

Further, the review delay will skew the results. You may find, as I
have, that Search Network gives a lower CTR than Google Search. You
will notice that a new ad starts out strong, and then the CTR drops
once Search Network kicks-in.

I've done some testing that I believe I've found a way to isolate
Google Search and Search Network CTRs. In the particular case of the ad
campaign that I am testing, Search Network appears to be giving about
1/2 the CTR of Google Search. Now, tell me how you can accurately
compare ads when there is a variable, unknown delay between when an ad
is created and when it is reviewed and starts running on Search
Network?

(One way to improve accuracy is to pause the campaign until the ad has
been reviewed. Of course, you have no way of knowing when it has been
reviewed in this case, other than calling a Google rep and bugging
them.)

If you don't pause the campaign, you have to wait until your ads are
serving 50/50, and then still wait a few more days before comparing
CTRs, and even then it is not perfect. It seems to me if you EVER turn
on Search Network, then you've made it impossible to accurately compare
CTRs. Part of the time your ad was running on Google Search only, and
part of the time it was running on Google Search and Search Network.
You don't even know exactly when the ad was reviewed and started
running on Search Network (unless a rep tells you).

So, you don't know how much your changes to your ad contributed to the
difference in CTR, and how much having the ad running on Search Network
once it was reviewed contributed to the difference.

This has led me astray, and I'm sure has led others astray. The net
result is that it can lead you to conclude that the inferior ad is
actually the better one! This is because it is the OLDER ad, and thus
has run more on Google Search than the newer ad.