the long answer is that my skepticism comes from climate being a classic complexity issue where trying to isolate a single factor (carbon) as being the main forcing for climate change requires a lot of real evidence.
i'd turn the question back around on you by asking "why do you believe in CAGW?" when i asked myself that question (because i was like most people and believed the standard news reports before i tried to really understand it), i discovered two main pieces of observational data that were overwhelming for why the increased carbon is going to lead to run-away warming:
1. the correlation of atmospheric carbon and temperature in the geologic record (ice core samples and tree ring data). it's very clear looking at those records that atmospheric carbon and temperature are correlated. but, when i dug into it i found out that in the geologic record the temperature increase PRECEEDS atmospheric CO2 by hundreds of years. if we were going to infer causation by correlation, then it would have to be the other way around. so that strong piece of evidence turned out not to be actual evidence.
when i sought answers for this from the global warming alarmist scientists, what they had to say was "yes, it's true. what we know is that the temperature increases for some unknown reason, then carbon is released because of the temperature, which then causes the temperature to increase more and a positive feedback loop is created."
but, if this were really how it happened and co2 is the overwhelmingly main forcing for temperature increase, then why wasn't the planet already destroyed back when we had 10x+ the amount of carbon in the atmosphere than what we are doing today?
2. after #1 melted away, i realized that the strongest evidence for CAGW was rooted in the predictions of climate models. it wasn't being proven by geologic or current observations, so what do the models tell us?
well, the models are very interesting because they are mechanisms we use to try to understand a classic complexity problem. these are notoriously difficult problems to fully understand down to it's constituent components.
add on top of it the timeframes we are talking about with the models, and it becomes very difficult to understand it right now. the models predict noticeable temperature increase in the 100s of years timeframe, not in the decadal timeframe.
so to be able to actually observe the predictions of the models, we'd have to wait 100s of years.
so, i wondered if there were other things that the models predicted that were more in the annual/decadal timeframe that we could observe to be happening.
as it happens, the answer is yes. all models that predicted runaway warming due to the co2 greenhouse effect ALSO predicted a warming of the atmosphere at a greater rate than a warming of the surface. (which makes sense based on the theory)
so all we'd have to do, is show that the atmosphere is heating up at an accelerated rate than the surface, and that would be a significant data point.
but what we actually observe is the opposite. the surface temperature is increasing at a faster rate than the atmospheric temperature. so the observed data appears to be going against the theory.
now, to be clear, i don't disagree that the planet is warming. i'm just skeptical that an increase in co2 is the main culprit.
with that in mind, i'd love to hear other people's observational evidence that backs the theory up. that's what will convince me.
and to make it clear, what will not convince me are fear mongering about "what if the theory is right?!" or the quasi-religious appeals to authority by saying "all these smart people believe different than you."
i want predictions made by the theory that show observational corroboration (with, preferably, an elimination of other explanations for the outcome).