I see many kayakers struggle with how to read and estimate the currents on the bay. This often leads to route decisions that seem strange or at least unnecessarily strenuous.
Understanding and predicting the current is far from an exact science, but there is a lot you can do to improve your odds and really try to get a grip on what the current is doing. I'll take this opportunity to offer a brief write-up of techniques I like to use to deal with this.
Before I started kayaking, I was racing sailboats very actively on the Bay. Like a kayak, a sailboat is heavily dependent on wise choice of paths, and to do well in sailboat racing on the Bay, you really have to know your currents well. My main role on the boat was that of tactician; to decide where to position the boat with respect to wind, current and the competition, in order to maximize the odds of winning. During the late 1990s and early 2000s I won or placed toward the top of several of the more competitive sailboat races on the bay. The task of judging the current is a little easier in a sailboat as the more elevated position standing on the deck of a sailboat makes it a bit easier to see what's happening over a larger area, but the techniques outlined here are highly applicable to kayaking as well.
To "predict what is going to happen" is possible most of the time given some diligent homework, and some basic understanding of how the currents work. More importantly though—there are tons of visual cues available to the trained eye. If you pay attention, and know what to look for, you can usually get a see what the current is doing at a distance.
Know your tides. Know the predicted time and strength for max current and the time of slack water at key locations. Know the time and height of the relevant high and low tides. [In addition, being aware of recent rainfall and associated unusual flow volumes in the Sacramento river also helps in the north and central bay.] Using this info, along with a watch to check time of day, you can get a good grip on roughly what to expect. With experience you will learn what is going on.
Momentum: Newton's first law of motion: "A body in motion stays in motion [...]". Think of each separate current as an enormous river where the massive body of water naturally tries to maintain its momentum and it takes a lot of energy to turn or divert that moving mass.
Path of least resistance: The current is stronger where it faces the least resistance. The current is stronger in deep water than in shallow water [do not confuse this with the reverse condition in a river of restricted width, where uniformly shallow sections have stronger current than uniformly deep sections do]. And the current is stronger away from shore than it is close to shore.
This combined with a grip on the bathymetry (topology under water) should give an idea of where to expect the main flow to occur, and where eddies might form.
The most important, and often the most underestimated and neglected, tool available for predicting the current is to watch the water carefully and pay attention to details. There are several types of cues that together can give you a very good view of what the current is doing at the moment. This combined with the proper preparation and basic physics understanding outlined above should give you far better odds than average to predict the current.
Boats: By observing how other boats are moving, you can judge what current they are in. Slow boats are better tell tales than fast boats, as they are affected more by the current, so sailboats, rowboats and kayaks are good targets. A sailboat with full sails that barely moves is likely fighting a current. Conversely, a boat that seems to move unnaturally fast is likely carried by a favorable current. And any boat moving sideways is clearly being set by a current.
Birds and other floating objects: If you look carefully you will likely find many objects floating in the bay that help you track the current that the casual observer might miss. There are usually many species of birds in the bay. They make excellent current indicators. Similarly, there is far too much debris around, like logs or various plastic objects, that can also be used as indicators.
Waves: You can use the shape and appearance of small waves to assess the current. A wave in water moving against the wind is steeper and stands up more (appears darker) than a wave in water that moves with the wind. With experience you can use this to identify different sections of water with different speeds relative to the wind. Care needs to be applied though, since there are several other factors that also impact the surface appearance, like varying wind speed, reflections of the background or sky with different color tones, etc.
Foam lines: Where currents meet, and between areas of current with different speeds, foam and floating debris often tends to line up. Where the speed difference between two currents is large, you can often see choppy water in a fairly well delineated area.
Combining all of the above provides a powerful way to understand and predict the current. Note that this isn't something that is perfected instantaneously, but by aggregating observations over minutes and hours as you paddle, you gradually build a dynamic map of what is happening on the water. Remember that it's a constantly changing environment, and add the basic tide info from the preparation, and an understanding of how bodies of water in motion behave, and the picture gets clearer. Over time accumulating the experiences, you can develop an ability to grok what the current is doing that goes far beyond what you can read in a tide log, or get from a casual glance of the water. Finally remember that this is not an exact science, and at times it all still can be completely wrong, and you'll look strange or select an unnecessarily strenuous route...