这个测试让我有点摸不透了啊, 谁能解释下呢?

代码如下:
def test(func):
@wraps(func)
def wrapper(*args, **kwargs):
t = time.time()
func(*args, **kwargs)
print func.__name__, time.time() - t
return
return wrapper


@test
def format_test():
for i in range(_item):
s = '{a}hello{b}hello{c}'.format(a='wen', b='guo', c='wen')


@test
def plus_test():
for i in range(_item):
s = 'wen'+'hello'+'guo'+'hello'+'wen'


@test
def tuple_test():
for i in range(_item):
s = '%shello%shello%s' % ('wen', 'guo', 'wen')


if __name__ == '__main__':
format_test()
plus_test()
tuple_test()

结果:
format_test 0.0439848899841
plus_test 0.00256514549255
tuple_test 0.0123269557953


问题:
为啥直接加花费的时间要低呢? 这不科学啊, 一直以为它应该是最慢的才对啊啊啊