Why does pymc.MAP not always return the same value - pymc

Question

I am running pymc2 to fit a straight line through my data. The code is shown below (modified from examples I found online). When I call the MAP function multiple times, I get different answers, even though I start with the exact same model. I thought the optimization method, fmin_powell, starts at the supplied value for each parameter. As far as I know, fmin_powell has no random component, so it should always end at the same optimum, yet it doesn't. Why do I keep getting different results?
import numpy as np
import pymc
# observed data
n = 21
a = 6
b = 2
sigma = 2
x = np.linspace(0, 1, n)
np.random.seed(1)
y_obs = a * x + b + np.random.normal(0, sigma, n)
def model():
# define priors
a = pymc.Normal('a', mu=0, tau=1 /10 ** 2, value=5)
b = pymc.Normal('b', mu=0, tau=1 / 10 ** 2, value=1)
tau = pymc.Gamma('tau', alpha=0.1, beta=0.1, value=1)
# define likelihood
#pymc.deterministic
def mu(a=a, b=b, x=x):
return a * x + b
y = pymc.Normal('y', mu=mu, tau=tau, value=y_obs, observed=True)
return locals()
ml = model() # dictionary of all locals
mcmc = pymc.Model(ml) # MCMC object
mapmcmc = pymc.MAP(mcmc)
mapmcmc.fit(method='fmin_powell')
print(mcmc.a.value, mcmc.b.value, mcmc.tau.value)
ml = model() # dictionary of all locals
mcmc = pymc.Model(ml) # MCMC object
mapmcmc = pymc.MAP(mcmc)
mapmcmc.fit(method='fmin_powell')
print(mcmc.a.value, mcmc.b.value, mcmc.tau.value)
ml = model() # dictionary of all locals
mcmc = pymc.Model(ml) # MCMC object
mapmcmc = pymc.MAP(mcmc)
mapmcmc.fit(method='fmin_powell')
print(mcmc.a.value, mcmc.b.value, mcmc.tau.value)