Phantom dark energy can produce amplified cosmic acceleration at late times, thus increasing the value of $H_0$ favored by CMB data and releasing the tension with local measurements of $H_0$. We show that the best fit value of $H_0$ in the context of the CMB power spectrum is degenerate with a constant equation of state parameter $w$, in accordance with the approximate effective linear equation $H_0 + 30.93\; w - 36.47 = 0$ ($H_0$ in $km \; sec^{-1} \; Mpc^{-1}$). This equation is derived by assuming that both $\Omega_{0 \rm m}h^2$ and $d_A=\int_0^{z_{rec}}\frac{dz}{H(z)}$ remain constant (for invariant CMB spectrum) and equal to their best fit Planck/$\Lambda$CDM values as $H_0$, $\Omega_{0 \rm m}$ and $w$ vary. For $w=-1$, this linear degeneracy equation leads to the best fit $H_0=67.4 \; km \; sec^{-1} \; Mpc^{-1}$ as expected. For $w=-1.22$ the corresponding predicted CMB best fit Hubble constant is $H_0=74 \; km \; sec^{-1} \; Mpc^{-1}$ which is identical with the value obtained by local distance ladder measurements while the best fit matter density parameter is predicted to decrease since $\Omega_{0 \rm m}h^2$ is fixed. We verify the above $H_0-w$ degeneracy equation by fitting a $w$CDM model with fixed values of $w$ to the Planck TT spectrum showing also that the quality of fit ($\chi^2$) is similar to that of $\Lambda$CDM. However, when including SnIa, BAO or growth data the quality of fit becomes worse than $\Lambda$CDM when $w< -1$. Finally, we generalize the $H_0-w(z)$ degeneracy equation for $w(z)=w_0+w_1\; z/(1+z)$ and identify analytically the full $w_0-w_1$ parameter region that leads to a best fit $H_0=74\; km \; sec^{-1} \; Mpc^{-1}$ in the context of the Planck CMB spectrum. This exploitation of $H_0-w(z)$ degeneracy can lead to immediate identification of all parameter values of a given $w(z)$ parametrization that can potentially resolve the $H_0$ tension.

A redshift tomography of the Pantheon type Ia supernovae (SnIa) data focusing on the best fit value of the absolute magnitude $M$ and/or Hubble constant $H_0$ in the context of $\Lambda$CDM indicates a local variation ($z\lesssim 0.2$) at $2\sigma$ level, with respect to the best fit of the full dataset. If this variation is physical, it can be interpreted either as a locally higher value of $H_0$, corresponding to a local matter underdensity $\delta \rho_0/\rho_0 \simeq -0.10 \pm 0.04$ or as a time variation of Newton's constant which implies an evolving Chandrasekhar mass and thus an evolving absolute magnitude $M$ of SnIa. The local void scenario would predict an anisotropy in the best fit value of $H_0$ since it is unlikely that we are located at the center of a local spherical underdensity. Using a hemisphere comparison method we find an anisotropy level consistent with simulated isotropic datasets. We show however, that the anisotropic sky distribution of the Pantheon SnIa data induces a preferred range of directions even in simulated Pantheon data obtained in the context of isotropic $\Lambda$CDM. We thus construct a more isotropically distributed subset of the Pantheon SnIa and show that the preferred range of directions disappears. Using this subset we again find no evidence for anisotropy using either the hemisphere comparison method or the dipole fit method. In the context of the modified gravity scenario, we allow for an evolving normalized Newton's constant consistent with General Relativity (GR) at early and late times $\mu(z)=1+g_a z^2/(1+z)^2-g_a z^4/(1+z)^4$ and fit for $g_a$ assuming $L\sim G_{\rm{eff}}^b$. For $b=-3/2$ indicated by previous studies we find $g_a=-0.47 \pm 0.36$ which is more than $1.5\sigma$ away from the GR value of $g_a=0$. This weak hint for weaker gravity at low $z$ is consistent with similar evidence from growth and weak lensing data.

Dynamical observational probes of the growth of density perturbations indicate that gravity may be getting weaker at low redshifts $z$. This evidence is at about $2-3\sigma$ level and comes mainly from weak lensing data that measure the parameter $S_8=\sigma_8 \sqrt{\Omega_{0m}/0.3}$ and redshift space distortion data that measure the growth rate times the amplitude of the linear power spectrum parameter $f\sigma_8 (z)$. The measured $f\sigma_8$ appears to be lower than the prediction of General Relativity (GR) in the context of the standard $\Lambda$CDM model as defined by the Planck best fit parameter values. This is the well known $f\sigma_8$ tension of $\Lambda$CDM, which constitutes one of the two main large scale challenges of the model along with the $H_0$ tension. We review the observational evidence that leads to the $f\sigma_8$ tension and discuss some theoretical implications. If this tension is not a systematic effect it may be an early hint of modified gravity with an evolving effective Newton's constant $G_{eff}$ and gravitational slip parameter $\eta$. We discuss such best fit parametrizations of $G_{eff}(z)$ and point out that they can not be reproduced by simple scalar-tensor and $f(R)$ modified gravity theories because these theories generically predict stronger gravity than General Relativity (GR) at low $z$ in the context of a $\Lambda$CDM background $H(z)$. Finally, we show weak evidence for an evolving reduced absolute magnitude of the SnIa of the Pantheon dataset at low redshifts ($z<0.1$) which may also be explained by a reduced strength of gravity and may help resolve the $H_0$ tension.

General Relativity (GR) is consistent with a wide range of experiments/observations from millimeter scales up to galactic scales and beyond. However, there are reasons to believe that GR may need to be modified because it includes singularities (it is an incomplete theory) and also it requires fine-tuning to explain the accelerating expansion of the universe through the cosmological constant. Thus, it is important to check various experiments and observations beyond the above range of scales for possible hints of deviations from the predictions of GR. If such hints are found it is important to understand which classes of modified gravity theories are consistent with them. The goal of this review is to summarize recent progress on these issues. On sub millimeter scales we show an analysis of the data of the Washington experiment (Kapner et al. (2007)) searching for modifications of Newton's Law on sub-millimeter scales and demonstrate that a spatially oscillating signal is hidden in this dataset. We show that even though this signal cannot be explained in the context of standard modified theories (viable scalar tensor and $f(R)$ theories), it is a rather generic prediction of nonlocal gravity theories. On cosmological scales we review recent analyses of Redshift Space Distortion data which measure the growth rate of cosmological perturbations at various redshifts and show that these data are in some tension with the $\Lambda$CDM parameter values indicated by Planck/2015 CMB data at about 3$\sigma$ level. This tension can be reduced by allowing for an evolution of the effective Newton constant that determines the growth rate of cosmological perturbations. We conclude that even though this tension between the data and the predictions of GR could be due to systematic/statistical uncertainties of the data, it could also constitute early hints of a new gravitational theory.

A cosmological observable measured in a range of redshifts can be used as a probe of a set of cosmological parameters. Given the cosmological observable and the cosmological parameter, there is an optimum range of redshifts where the observable can constrain the parameter in the most effective manner. For other redshift ranges the observable values may be degenerate with respect to the cosmological parameter values and thus inefficient in constraining the given parameter. These are blind redshift ranges. We determine the optimum and the blind redshift ranges of cosmological observables with respect to the cosmological parameters: matter density parameter $\Omega_m$, equation of state parameter $w$ and a modified gravity parameter $g_a$ which parametrizes the evolution of an effective Newton's constant. We consider the observables: growth rate of matter density perturbations expressed through $f(z)$ and $f\sigma_8$, the distance modulus $\mu(z)$, Baryon Acoustic Oscillation observables $D_V(z) \times \frac{r_s^{fid}}{r_s}$, $H \times \frac{r_s}{r_s^{fid}}$ and $D_A \times \frac{r_s^{fid}}{r_s}$, $H(z)$ measurements and the gravitational wave luminosity distance. We introduce a new statistic $S_P^O(z)\equiv \frac{\Delta O}{\Delta P}(z) \cdot V_{eff}^{1/2}$, including the effective survey volume $V_{eff}$, as a measure of the constraining power of a given observable $O$ with respect to a cosmological parameter $P$ as a function of redshift $z$. We find blind redshift spots $z_b$ ($S_P^O(z_b)\simeq 0$) and optimal redshift spots $z_s$ ($S_P^O(z_s)\simeq max$) for these observables with respect to the parameters $\Omega_m$, $w$ and $g_a$. For $O=f\sigma_8$ and $P=(\Omega_{m},w,g_a)$ we find blind spots at $z_b\simeq(1,2,2.7)$ respectively and optimal (sweet) spots at $z_s=(0.5,0.8,1.2)$. Thus probing higher redshifts may be less effective than probing lower redshifts with higher accuracy.

Recent analyses \cite{Nesseris:2017vor,Kazantzidis:2018rnb} have indicated that an effective Newton's constant $G_{\rm eff}(z)$ decreasing with redshift may relieve the observed tension between the Planck15 best fit $\Lambda$CDM cosmological background ({\it i.e.} Planck15/$\Lambda$CDM) and the corresponding $\Lambda$CDM background favored by growth $f\sigma_8$ and weak lensing data. We investigate the consistency of such a decreasing $G_{\rm eff}(z)$ with some viable scalar-tensor models and $f(R)$ theories. We stress that $f(R)$ theories generically can not lead to a decreasing $G_{\rm eff}(z)$ for any cosmological background. For scalar-tensor models we deduce that in the context of a $\Lambda$CDM cosmological background, a decreasing $G_{\rm eff}(z)$ is not consistent with a large Brans-Dicke parameter $\omega_{BD,0}$ today. This inconsistency remains and amplifies in the presence of a phantom dark energy equation of state parameter ($w < -1$). However it can be avoided for $w >-1$. We also find that any modified gravity model with the required decreasing $G_{\rm eff}(z)$ and $G_{{\rm eff},0}=G$, would have a characteristic signature in its growth index $\gamma$ with $0.61\lesssim \gamma_0\lesssim 0.69$ and large slopes $\gamma_0'$, $0.16\lesssim \gamma_0'\lesssim 0.4$, which is a characteristic signature of a decreasing (with $z$) $G_{\rm eff}(z)<G$ on small redshifts. This is a substantial departure today from the quasi-static behaviour in $\Lambda$CDM with $(\gamma_0,\gamma_0')\approx (0.55,-0.02)$.

We construct an updated extended compilation of distinct (but possibly correlated) $f\sigma_8(z)$ Redshift Space Distortion (RSD) data published between 2006 and 2018. It consists of 63 datapoints and is significantly larger than previously used similar datasets. After fiducial model correction we obtain the best fit $\Omega_{0m}-\sigma_8$ $\Lambda$CDM parameters and show that they are at a $5\sigma$ tension with the corresponding Planck15/$\Lambda$CDM values. Introducing a nontrivial covariance matrix correlating randomly $20\%$ of the RSD datapoints has no significant effect on the above tension level. We show that the tension disappears (becomes less than $1\sigma$) when a subsample of the 20 most recently published data is used. A partial cause for this reduced tension is the fact that more recent data tend to probe higher redshifts (with higher errorbars) where there is degeneracy among different models due to matter domination. Allowing for a nontrivial evolution of the effective Newton's constant as $G_{\textrm{eff}}(z)/G_{\textrm{N}} = 1 + g_a \left(\frac{z}{1+z}\right)^2 - g_a \left(\frac{z}{1+z}\right)^4$ ($g_a$ is a parameter) and fixing a \plcdm background we find $g_a=-0.91\pm 0.17$ from the full $f\sigma_8$ dataset while the 20 earliest and 20 latest datapoints imply $g_a=-1.28^{+0.28}_{-0.26}$ and $g_a=-0.43^{+0.46}_{-0.41}$ respectively. Thus, the more recent $f\sigma_8$ data appear to favor GR in contrast to earlier data. Finally, we show that the parametrization $f\sigma_8(z)=\lambda \sigma_8 \Omega(z)^\gamma /(1+z)^\beta$ provides an excellent fit to the solution of the growth equation for both GR ($g_a=0$) and modified gravity ($g_a\neq 0$).