Did you know that there are almost 2000 ETFs on the market today? Analysts estimate that we will average 150+ new ETF's a year for the next 4 years. So with all these choices it can be tough to determine which product will work best in your portfolio. How should you evaluate the ever-expanding ETF landscape?

Start with what's under the hood -

People always like to focus on the ETF's fee's, or its size, and  all those things matter, but to me, the single most important thing to consider about an ETF is its underlying index. 

We're conditioned to believe that all indexes are the same. The S&P 500 and the Russell 1000: What's the difference?

The answer is: Not much. Sure, the Russell 1000 has twice as many securities as the S&P 500. But over any given period, the two will perform about the same. Moreover, who's to say one will be up more than the other?

But outside of a few cases, indexes matter ... a lot. The Dow Jones industrial average holds 30 stocks, and looks (and performs) nothing like the S&P 500. One popular China ETF tracks an index that's 50 percent financials; another tracks an index with no financials at all.

One of the beautiful things about ETFs is that they disclose their holdings (mostly) on a daily basis. So take the time to look under the hood and see if the holdings, sector and country breakdowns make sense. Do they match the asset allocation you have in mind?

Pay particular attention not just to what stocks or bonds an ETF holds, but how they're weighted. Some indexes weight their holdings more or less equally, while others allow one or two big names to shoulder the burden. Some aim for broad market exposure, while others take risks in an attempt to outperform the market.  Just because you are buying one basket of stocks that replicates the S&P 500 doesn't mean you should't assume that all ETFs are the same, because they definitely aren't!. 

How High Is Its Tracking Difference?

Most people will never think to look at this. Once you've found the right index, it's important to make sure the fund is reasonably priced, well-run and tradable. Most investors start with a fund's expense ratio  assuming the lower the better. 

But expense ratios aren't the be-all-end-all. You should take a moment to look at a fund's "tracking difference." ETF's are designed to track indexes. If an index is up 10.25 percent, a fund should be up 10.25 percent too. But that's rarely the case. 

First, expenses create a drag on returns. If you charge 0.25 percent in annual fees, your expected return will be 10.00 percent (10.25% - 0.25% in annual fees). But beyond expenses, some issuers do a better job tracking indexes than others. Also, some indexes are easier to track than others.

Let's start with the base case. For a popular large-cap U.S. equity index like the S&P 500, most ETF's tracking that fund will use what's called "full replication." That means they buy every security in the S&P 500 at the exact ratio at which they are represented in the index. Before transaction costs, this fund should track the index perfectly.

But what if they are tracking an index in Vietnam that has a lot of turnover? Transaction costs can eat away into returns.

Sometimes, fund managers will only buy some of the stocks or bonds in an index. This is called "sampling," or more optimistically, "optimization." A sampled strategy will typically aim to replicate an index, but it may over perform, or under perform slightly based on what securities it holds.

All in all, the lower the tracking difference is, especially on the downside, the better. 

How Liquid Is The Fund?

If a fund has the right strategy and is well-run, you then decide if you can buy it. Trading costs can eat into your returns if you're not careful. 

The three things you want to look for are the fund's liquidity; its bid/ask spread; and its tendency to trade in line with its true net asset value.

An ETF's liquidity stems from two sources, the liquidity of the fund itself; and the liquidity of its underlying shares. Funds with higher average daily trading volumes and more assets under management tend to trade at tighter spreads than funds with less daily trading or lower assets. But even funds with limited trading volume can trade at tight spreads if the underlying securities of the fund are liquid. An ETF that invests in S&P 500 stocks, for example, will probably be more liquid than one that invests in Brazilian small-caps or alternative energy companies. It just makes sense.