This is gonna be pretty long-winded, but here goes...
You could put a bar and chain on the saw to see if it may idle down more normally. Otherwise, it could indeed be an air leak which can exhibit several symptoms depending on severity. Not decelerating smoothly to a normal idle is one. Not idling or wandering at idle is another. Revving or stalling when the saw is turned on its side is another. Having to choke it when starting even when warm is another. Not being able to get a decent tune is probably the most frustrating and most common, though often written off as a carb problem until a pressure/vacuum test ultimately reveals an air leak.
Generally speaking, if you have any two (or more) of the above symptoms you will find an air leak somewhere in the intake assembly or crankcase..., which includes the cylinder base gasket, loose cylinder bolts, or even a cracked cylinder base or case.
A redneck leak check in the absence of a proper vac/pressure test rig is to spray carb cleaner (or something similar) at suspected areas like the intake boot/impulse hose and behind the flywheel and clutch (where such access to the crank seal area is safely obtained) while the saw is running to see if it changes behavior by sucking in whatever you're spraying. If it does, you probably have a fairly large leak in that particular area.
A vacuum/pressure test will determine the integrity of the crankcase gasket, base gasket/cylinder seal, oil seals, and whatever else may be bolted on with threads that go all the way into the case (like oil pumps). It can also test the entire intake assembly at the same time on most saws depending on one's setup and the intake design. (The test point can be the impulse location/hose, an intake adapter/blockoff plate with a small hose barb in place of the carburetor, or a sparkplug adapter.)
Other sources of air leaks can obviously be loose spark plugs and decompression valves, but they rarely exhibit the same symptoms as intake or case leaks and are usually discovered fairly quickly simply by a saw's lack of a compression. While not having the same type of performance impact, they will still need to be accounted for in a vac/pressure test and corrected if possible. They are often overlooked as leak sources while oil seals and the intake area receive all the attention after a leak is actually discovered during a test. So when the soapy water starts getting sprayed all over the place looking for bubbles, I usually hit the plug and decomp first just to rule them out or to see how much they may be actually be leaking. It's not uncommon to find very small leaks at the plug and decomp and they're usually nothing to be overly concerned about. It's just good to know they're there as you proceed to the other major areas of concern so they can either be factored in or out of the overall results.
In the absence of vacuum test capability where only a pressure source is available, many will rely solely on a pressure test to determine if a leak exists. While pressure testing is part of the overall leak down check process, it should never be relied on to rule out the possibility of a vacuum leak for one very important reason: pressure can close a leak that only occurs under a vacuum. This is typical with bad oil seals in particular. Think of a faulty rubber seal around a crankshaft as a one-way valve that only allows air into the crank case, but doesn't let air out..., kinda like a duckbill valve works. Put pressure into the case and the valve closes. Fire up the saw and case vacuum opens it right back up again. While a bit more uncommon, the opposite can also occur. Always test both vac and pressure to ensure the overall air tight integrity of a saw whenever possible.
And while applying pressure and using soapy water is almost always the first approach to locating vacuum leaks, the above can also be applied to this testing method. There is definitely a vacuum leak somewhere, but the saw holds pressure. No bubbles. Most will simply apply more pressure figuring it will expose the leak sooner or later. This generally just reinforces the seal already obtained by the original amount of pressure, or in extreme cases, could blow out what is otherwise a perfectly good seal to begin with. So what do ya do? Use less pressure..., as in barely bumping the needle on the gauge. The example here would be akin to trying to inflate a tire on a rim without enough pressure to seal the bead. The air easily escapes. Same holds true for the most elusive vac leaks which seal right up with too much pressure.
Another method many use for those suspected little crank seal vac leaks that disappear with pressure is to apply some grease or heavy oil to the seal while applying vacuum. A small vac leak will often times be temporarily sealed by the grease/oil with the result being easily observed on a test gauge. In cases where a leak persists but is notably slower, the grease or oil can usually be seen getting drawn into the seal at the point of the leak. (When doing one of these tests make sure the area has been cleaned up and dried from any previous soapy water tests so as not to suck any residual moisture into the case.)
In a nutshell, air leaks are generally fairly easy to pinpoint by using simple methods and a decent vac/pressure setup. Many use the MityVac 8500 which can be had for around $55 these days if you shop around. All metal. All serviceable parts. Does both vacuum and pressure at the flip of a knob/valve. Originally for one-man brake bleeding but couldn't have been designed better for chainsaw/2 stroke engine use if they tried. Pays for itself the first time it finds a leak that woulda otherwise burned up your favorite saw (or someone else's).
Wouldn't be without mine.
