As every PDS disclosure says, "Past Performance Does Not Guarantee Future Results."
I often start (on the contrary) by looking at areas which suit certain criteria and have not necessarily performed the best.
When looking to value a house in a suburb there is 2 prices which move independently in the short term, but on the long term they converge. The first price is the market price (short term), which is determined by market sentiment – mainly buyers. This is what statistics measure. The second price is the intrinsic value or what it could be worth when you are looking to sell it – long term.
A great suburb, for me, is one where the intrinsic value of the place has increased (government spending has risen, crime rates are falling, private investment is moving in, zoning changes, and so on) and the market hasn't yet recognised this through price growth. The reason for this is the minute the market realises these qualities the price growth will follow.
Also, have a look into the statistic flaws of median house price. Without knowing where you got those percentages, I am guessing a lot of those stats would be calculated of median growth.