I have a pretty sketchy understanding of immunology, but my impression was that IgM antibodies represent a recent production, whilst IgG is more long term, so that early on in the process you'll see higher IgM levels, but later the IgG levels will rise. I don't think that you can draw conclusions about higher or lower risk of thrombosis depending on which type of antibody it is. There is some evidence that the higer the number of antibodies the higher the risk of clotting, but even this isn't always true - there are plenty of people out there who have ACl antibodies who have never had a clinical problem, and also those who have clinical APS but no antibodies.
Personally I've only had low positives for ACl antibodies (twice), and they were IgG type both times (normal IgM). Still I have clinical problems (5 miscarriages and 3 episodes of thromboplebitis).
For some reason, my ob-gyn thought that IgG antibodies were "risker" in terms of miscarriage & pregnancy problems than IgM anti-cardiolipin antibodies. The research I did at the time was mixed - some studies agreed with his statement and other studies found significant increased risk with either type of ACL antibody. Most studies felt higher levels indicated more risk.
I have tested positive for ACL IgM twice (a few months apart), once was in the mid 20s and once in the 50s. It would be interesting to see what has happened now over 5 years later - if they've changed to the IgG type. I've only had thrombophelebitis once, and no other problems due to them that I am aware of.