ihate17 was telling it like it is, far as how it goes trying this sort of thing. you really if your going to try and do this stuff need to try to keep a good handle on the previous rounds.
so but what your describing is a lot like what i've been trying to do. like i counted the orthodox way for a few years or so and still do some times. i'd just say if you want to try and do a 'fuzzy' count sort of thing that having had some experience as an orthodox counter might be a help. you at least then have an idea of how it goes in an experiential sort of way.
one thing that i do is practice one heck of a lot with cvbj. it's good because you can go back and examine your cvbj logs and see how well your bets correlate with the actual true counts. i can tell you, that at least in my case i'm pretty darn far off count wise from what i think the count should be and from what the true count actually is. unfortunately i find that it's not all that infrequent that i've been betting up into a negative count

.
so, but anyway, what i try to do is i'll run a simulation for the particular game i'm going to play. i'll determine the standard deviation, expectation and probable range of results for a given number of hands that a perfect card counter can expect betting a certain way so as to attain the risk of ruin and expectation that is acceptable to me. then i'll play my fuzzy count voodoo stuff trying my best to mimic the play of that perfect card counter with out really counting but just mainly intuitively trying to keep a handle on how rich or poor the pack to be dealt is in high cards.
probably i guess stop loss play, and having goal targets is something that most orthodox counters would ignore. well, imho a 'fuzzy' counter should have stop loss's and goal targets. like in my case, say maybe i lose about one standard deviation worth of units over some number of hands, i might just abandon the 'fuzzy' stuff and try and count orthodox, alternatively if i manage to reach expectation or one standard deviation to the good i'll probably quit for the day or at least stop play for a while since i realize i've just some how managed to do as well for the time being as a perfect card counter, lol.
one thing that can help you keep a handle on how the count may be going is by keeping track of situational hands in combination with 'leaning'.
leaning is just the sort of thing you was describing, watching out for those time when a lot of low cards present, while hopefully keeping a mental note of how the previous rounds went. situational hands are the particular types of hands that you, the dealer and other players may have in a given round. when you and or the other players are doing a lot of splitting, splitting and then doubling down or just doubling down on hard hands, then it likely that the count is rising. splitting aces is usually not good news for the count. doubling soft hands is not such good news as making hard doubles. to a lesser extent it is also likely that the count is rising when you and the dealer or the other players and the dealer have a number of cards that add up to seven or more cards.
well all this said, lmao, don't try this stuff in the casino, do it at home on cvbj.

oh and where you was hoping to make $200 or so on a consistent basis, well this wouldn't be the way. me so far i've been just breaking even after a year or so.