For most people, playing bridge online is synonymous with BBO. I like BBO too, but there is a new game in town and if you have not tried it out, you should.
Every day, bridgez.net runs a matchpointed bridge tournament. Although you can play online, the website is experiencing growing pains, so a much better avenue is to download the bridge software wbridge5 and play the bridgez tournament using it.
From what I can tell, the wbridge5 robot makes killer leads and messes up defensive signals just as often as BBO's GIB program. Its card-play and bidding are uncanny. However, when it is your partner (rather than your opponent) the bidding is a lot stranger. The software provides an explanation of sorts, in terms of high card points and distributions, but that explanation is often quite wrong. For example, although the robot marked my fourth suit bid as artificial (0-4 hearts), it had no problems raising me to game in hearts. However, everyone else playing the tournaments is having the same problems (everyone plays 2/1), so I think of it as playing with a slightly unpredictable partner. Which is, of course, always the case.
An advantage of long-running tournaments like this is the size of the field -- every day, the tournament attracts about 500 users, so you have a large field to gauge things against. For example, you can experiment with different styles to see what works. Thus, I have been trying more speculative matchpoint doubles. At the end of a hand, you can also click on how someone else played the hand to see why you dropped a trick.
Another very good thing about playing the bridgez tourneys is the monthly reports you get at the end of the month. For example, my monthly report at the end of June told me where I ranked in total ability:
It also analyzed my bidding in a rather cool manner:
The y-axis ("efficiency") indicates how good I am at analyzing the playing strength of both hands and bidding the right contracts. I am not sure what the x-axis ("frequency") denotes -- it is explained as "mastery of bidding system", so perhaps the below-50% reflects the times the software leaves me in my control-bids and fails to take preference to my first bid suit.
Yup, I am in total denial about my bidding ability.